Ratings - please read before commenting - Archived

[mattyp151]

Hi Mr Cat.

Indeed it was Matty who missed the coke vs pepsi point. I take that back. My apologies.

The lucky punch stuff is actually pretty irrelevant. Firstly no champ fights a 1-15 fighter. Secondly, if an upset does occur, then it usually an indication that further upsets will occur. Lennox got nailed by McCall and then Rahman. Wlad got beat up by a strong chinned Purrity and then Sanders and Brewster. Zab was an accident waiting to happen. He proved his loss to Spinks was no fluke, because he lost to Baldomir. I can't for the life of me think of a recent top 10 boxer who near the peak of his career lost via lucky punch and then went on to never lose again.

Martin actually knows the answer to this question, he mentioned it in his only response to me. It's all about levels i.e. the perceived level of a fighter. Is the figher an elite p4p, a champion, a serious contendor, a fringe contendor, a journeyman or a tomato can. If an elite fighter loses to 1-15 fighter, then he was never an elite fighter. he should be reduced to a champion level ( a poor one at that) or even a serous contendor. Conversely the journeyman should be elevated to the (poor) champion level or serious contendor level. Amongst this level, then A should be ranked ahead of B, at least until further data is available.

If there are several eilte fighters in a division, then then the former elite fighter should land behind them. By definition, anyone who gets KO'd by a bum is not an elite fighter. I did not say or imply, that the lucky puncher should inherit the previous spot occupied by the elite fighter. If I personally came out and KO'd Manny Pacquaio on my pro debut, then there is no way I should be ranked above Barerra, Raheem, Moralles etc.

The WBo rankings suck. What your mean is the IBO rankings. The IBO rankings rock, when they are up to date that is. Their algorithm is not public property, but they have the concept of 'levels' in it. All you have to do is to look at the points for a fighter and you can guage at what level they are at. You look at boxrec's and there is very little difference between 1 and 20.

You can be rest assured it's an algorithm though.

conan

I didn't misunderstand your soda model at all...it's a shit example. You compared a poll of two options to a poll of 15,000.

Put it this way, for your plausibility clause or whatever, a p4p list completely derails any theory of plausibility. Most p4p lists have Mayweather #1 or #2. You ever think he would have a chance to beat Nicolay Valuev? No way. So, why isn't Valuev #1? It's not plausible that Mayweather could beat anyone minimum to heavy, so thus, by your reason, the only #1 fighter could be a LHW, CW or HW pretty much.

You got a real short memory Matty, or you didnt read my response. I used Coke vs Pepsi as an example of how it is important to do a plausability check. I did not compare it as a poll between two options and a poll of 15000. Read the first post again and come back.

And who is talking about the p4p list? What has that got to do with the price of fish in China. Nothing. I am criticizing the regular lists, not the p4p list. I dont even care about the p4p list. What do you mean then by "by your reason". I haven't cited anything about the p4p list. Don't let your anger at me get you all confused and quote things that I never said.

conan

You're just arrogant is all I'm walking away with here. You seem to completely be missing that it is not a ranking system, but a prediction system. Is it plausible Calvin Brock could beat Valuev? Yes, Larry Donalad beat Vaulev, he's far from a top notch heavy. You keep bringing up odd examples like Butterbean, and these guys are nowhere near the top. Butterbean is 197, Bowe isn't even rated, and those seem to be your big hang ups.

Look, stop looking at it as a ranking. It's not a ranking, its a prediction scale. The system says taking into account that Brock could possibly beat Valuev, ONLY IF VALUEV IS NOT FIGHTING IN GERMANY. If Nicolay fought Calvin in Germany, Valuev wins, just as if Wlad fought Brock in Germany. It's a tough call on the US because Wlad has won here as well, but I wouldn't say it's out of the question Wlad could lose to Brock given stamina and chin issues. You keep on citing HW issues, when EVERYONE knows HW boxing is a disaster since Lennox retired, and even got worse when Vitali left. You picking at the weak chink, instead of looking at the rest.

CW - biggest issue is Cunningham at 14, but look at his body of work, he hasn't truly beaten a top 5 fighter outside of Jones, whom he barely beat. Jones' biggest fight? A draw vs Nelson or the KO over Braithwaite. If they fight in the US, Cunningham gets the nod over Jones for the 125 point hometown fight. If they fight neutral or in Panama, Jones gets the nod, which isn't out of the question seeing Cunningham hasn't had a big win since Jones, and Jones only lost a close decision.

LHW - Yes, Chad Dawson at 1, and Adamek at 8 may irk some people, I think Adamek and Dawson should both be 1 based on the fights I've seen. Both are tall rangy fighters, who throw at a good pace and have proven they can KO big fighters and also work their way to a decision. Drews is an anomaly until you realize his 1 loss was in Australia, and other than that, he's been out of Germany twice, once in his home Croatia, the other his first career bout in Hungary. Think of the hometown decisions coming out of Germany. You think Drews gets those deductions vs Briggs at home? Nope, and that 115-109 and 113-109 could easily swing the other way with a hometown gift of a round or 2 if you throw out the deductions.

SMW- Calzaghe and Kessler are universally #1 and 2 in that division. Expect Kessler to separate himself further after he beats Beyer.

MW - Winky 1, Taylor a very close 2, Abraham 3, and gets the hometown rating over those two. Plausible? Certainly. Very talented kid who fights in the boxing robbery capital, why shouldn't you be able to give Abraham the nod over Winky or Jermain?

Just a few examples of what is right, instead of harping on the mess that is the HW division.

[JCS]

I didn't misunderstand your soda model at all...it's a shit example. You compared a poll of two options to a poll of 15,000.

Put it this way, for your plausibility clause or whatever, a p4p list completely derails any theory of plausibility. Most p4p lists have Mayweather #1 or #2. You ever think he would have a chance to beat Nicolay Valuev? No way. So, why isn't Valuev #1? It's not plausible that Mayweather could beat anyone minimum to heavy, so thus, by your reason, the only #1 fighter could be a LHW, CW or HW pretty much.

You got a real short memory Matty, or you didnt read my response. I used Coke vs Pepsi as an example of how it is important to do a plausability check. I did not compare it as a poll between two options and a poll of 15000. Read the first post again and come back.

And who is talking about the p4p list? What has that got to do with the price of fish in China. Nothing. I am criticizing the regular lists, not the p4p list. I dont even care about the p4p list. What do you mean then by "by your reason". I haven't cited anything about the p4p list. Don't let your anger at me get you all confused and quote things that I never said.

conan

You're just arrogant is all I'm walking away with here. You seem to completely be missing that it is not a ranking system, but a prediction system. Is it plausible Calvin Brock could beat Valuev? Yes, Larry Donalad beat Vaulev, he's far from a top notch heavy. You keep bringing up odd examples like Butterbean, and these guys are nowhere near the top. Butterbean is 197, Bowe isn't even rated, and those seem to be your big hang ups.

Look, stop looking at it as a ranking. It's not a ranking, its a prediction scale. The system says taking into account that Brock could possibly beat Valuev, ONLY IF VALUEV IS NOT FIGHTING IN GERMANY. If Nicolay fought Calvin in Germany, Valuev wins, just as if Wlad fought Brock in Germany. It's a tough call on the US because Wlad has won here as well, but I wouldn't say it's out of the question Wlad could lose to Brock given stamina and chin issues. You keep on citing HW issues, when EVERYONE knows HW boxing is a disaster since Lennox retired, and even got worse when Vitali left. You picking at the weak chink, instead of looking at the rest.

CW - biggest issue is Cunningham at 14, but look at his body of work, he hasn't truly beaten a top 5 fighter outside of Jones, whom he barely beat. Jones' biggest fight? A draw vs Nelson or the KO over Braithwaite. If they fight in the US, Cunningham gets the nod over Jones for the 125 point hometown fight. If they fight neutral or in Panama, Jones gets the nod, which isn't out of the question seeing Cunningham hasn't had a big win since Jones, and Jones only lost a close decision.

LHW - Yes, Chad Dawson at 1, and Adamek at 8 may irk some people, I think Adamek and Dawson should both be 1 based on the fights I've seen. Both are tall rangy fighters, who throw at a good pace and have proven they can KO big fighters and also work their way to a decision. Drews is an anomaly until you realize his 1 loss was in Australia, and other than that, he's been out of Germany twice, once in his home Croatia, the other his first career bout in Hungary. Think of the hometown decisions coming out of Germany. You think Drews gets those deductions vs Briggs at home? Nope, and that 115-112 and 113-109 could easily swing the other way with a hometown gift of a round or 2 if you throw out the deductions.

SMW- Calzaghe and Kessler are universally #1 and 2 in that division. Expect Kessler to separate himself further after he beats Beyer.

MW - Winky 1, Taylor a very close 2, Abraham 3, and gets the hometown rating over those two. Plausible? Certainly. Very talented kid who fights in the boxing robbery capital, why shouldn't you be able to give Abraham the nod over Winky or Jermain?

Just a few examples of what is right, instead of harping on the mess that is the HW division.

Here fornicating here..

Also, I believe the latest revision has Taylor at #1 in 160..Also John at #1 -126 Good or bad?? You decide..

[conan_the_cribber]

You're just arrogant is all I'm walking away with here. You seem to completely be missing that it is not a ranking system, but a prediction system. Is it plausible Calvin Brock could beat Valuev? Yes, Larry Donalad beat Vaulev, he's far from a top notch heavy. You keep bringing up odd examples like Butterbean, and these guys are nowhere near the top. Butterbean is 197, Bowe isn't even rated, and those seem to be your big hang ups.

Look, stop looking at it as a ranking. It's not a ranking, its a prediction scale. The system says taking into account that Brock could possibly beat Valuev, ONLY IF VALUEV IS NOT FIGHTING IN GERMANY. If Nicolay fought Calvin in Germany, Valuev wins, just as if Wlad fought Brock in Germany. It's a tough call on the US because Wlad has won here as well, but I wouldn't say it's out of the question Wlad could lose to Brock given stamina and chin issues. You keep on citing HW issues, when EVERYONE knows HW boxing is a disaster since Lennox retired, and even got worse when Vitali left. You picking at the weak chink, instead of looking at the rest.

CW - biggest issue is Cunningham at 14, but look at his body of work, he hasn't truly beaten a top 5 fighter outside of Jones, whom he barely beat. Jones' biggest fight? A draw vs Nelson or the KO over Braithwaite. If they fight in the US, Cunningham gets the nod over Jones for the 125 point hometown fight. If they fight neutral or in Panama, Jones gets the nod, which isn't out of the question seeing Cunningham hasn't had a big win since Jones, and Jones only lost a close decision.

LHW - Yes, Chad Dawson at 1, and Adamek at 8 may irk some people, I think Adamek and Dawson should both be 1 based on the fights I've seen. Both are tall rangy fighters, who throw at a good pace and have proven they can KO big fighters and also work their way to a decision. Drews is an anomaly until you realize his 1 loss was in Australia, and other than that, he's been out of Germany twice, once in his home Croatia, the other his first career bout in Hungary. Think of the hometown decisions coming out of Germany. You think Drews gets those deductions vs Briggs at home? Nope, and that 115-109 and 113-109 could easily swing the other way with a hometown gift of a round or 2 if you throw out the deductions.

SMW- Calzaghe and Kessler are universally #1 and 2 in that division. Expect Kessler to separate himself further after he beats Beyer.

MW - Winky 1, Taylor a very close 2, Abraham 3, and gets the hometown rating over those two. Plausible? Certainly. Very talented kid who fights in the boxing robbery capital, why shouldn't you be able to give Abraham the nod over Winky or Jermain?

Just a few examples of what is right, instead of harping on the mess that is the HW division.

If I point out where you misunderstood me, how does that make me arrogant?

Of course its a ranking system. The fighters end up with a number next to them and they are presented in a sorted fashion. If that is not a ranking system, then what is?

You can use this ranking system for predictions, just like the ELO from FIDE. But it is still a ranking system.

You justify the ranking system, by saying it's great for predicting. I say, comared to what. Read my position summary. Try and understand it. It explains precisely why I mention Bowe at several places.

Yes there are some right bits. But my, there are an awful lot of big mistakes in there. Prominent ones, real easy to find. That was my analogy with the car. You cant say the car is ok because some bits are working.

conan

[mattyp151]

You're just arrogant is all I'm walking away with here. You seem to completely be missing that it is not a ranking system, but a prediction system. Is it plausible Calvin Brock could beat Valuev? Yes, Larry Donalad beat Vaulev, he's far from a top notch heavy. You keep bringing up odd examples like Butterbean, and these guys are nowhere near the top. Butterbean is 197, Bowe isn't even rated, and those seem to be your big hang ups.

Look, stop looking at it as a ranking. It's not a ranking, its a prediction scale. The system says taking into account that Brock could possibly beat Valuev, ONLY IF VALUEV IS NOT FIGHTING IN GERMANY. If Nicolay fought Calvin in Germany, Valuev wins, just as if Wlad fought Brock in Germany. It's a tough call on the US because Wlad has won here as well, but I wouldn't say it's out of the question Wlad could lose to Brock given stamina and chin issues. You keep on citing HW issues, when EVERYONE knows HW boxing is a disaster since Lennox retired, and even got worse when Vitali left. You picking at the weak chink, instead of looking at the rest.

CW - biggest issue is Cunningham at 14, but look at his body of work, he hasn't truly beaten a top 5 fighter outside of Jones, whom he barely beat. Jones' biggest fight? A draw vs Nelson or the KO over Braithwaite. If they fight in the US, Cunningham gets the nod over Jones for the 125 point hometown fight. If they fight neutral or in Panama, Jones gets the nod, which isn't out of the question seeing Cunningham hasn't had a big win since Jones, and Jones only lost a close decision.

LHW - Yes, Chad Dawson at 1, and Adamek at 8 may irk some people, I think Adamek and Dawson should both be 1 based on the fights I've seen. Both are tall rangy fighters, who throw at a good pace and have proven they can KO big fighters and also work their way to a decision. Drews is an anomaly until you realize his 1 loss was in Australia, and other than that, he's been out of Germany twice, once in his home Croatia, the other his first career bout in Hungary. Think of the hometown decisions coming out of Germany. You think Drews gets those deductions vs Briggs at home? Nope, and that 115-109 and 113-109 could easily swing the other way with a hometown gift of a round or 2 if you throw out the deductions.

SMW- Calzaghe and Kessler are universally #1 and 2 in that division. Expect Kessler to separate himself further after he beats Beyer.

MW - Winky 1, Taylor a very close 2, Abraham 3, and gets the hometown rating over those two. Plausible? Certainly. Very talented kid who fights in the boxing robbery capital, why shouldn't you be able to give Abraham the nod over Winky or Jermain?

Just a few examples of what is right, instead of harping on the mess that is the HW division.

If I point out where you misunderstood me, how does that make me arrogant?

Of course its a ranking system. The fighters end up with a number next to them and they are presented in a sorted fashion. If that is not a ranking system, then what is?

You can use this ranking system for predictions, just like the ELO from FIDE. But it is still a ranking system.

You justify the ranking system, by saying it's great for predicting. I say, comared to what. Read my position summary. Try and understand it. It explains precisely why I mention Bowe at several places.

Yes there are some right bits. But my, there are an awful lot of big mistakes in there. Prominent ones, real easy to find. That was my analogy with the car. You cant say the car is ok because some bits are working.

conan

Well, put some "glaring issues" out there.

[conan_the_cribber]

Well, put some "glaring issues" out there.

Hi Matty,

I'm not sure I understand what you mean by glaring issues. I didn't use that term. But if you mean concrete examples of where the rankings are wrong, then I think I've done that already. But to recap.

The long post regarding the light heavy division.

In the heavies, Calvin Brock at 1. Tony Thompson (my god) at 5 after beating everyone's bitch Dom Guinn? Shannon Briggs at 6? Have a look at the fighters that Briggs has been beating. Surface level OKish, but then look at the fighter's they've beaten. Briggs has been brilliant, he has been padding the record with fighters who have obscenely been padding their records.

At cruiser, the most prominent example is Mormeck at 12. That is such a crime. There are also several inflated records in the top 10.

Super Middles has Beyer at 12?? But at least its got Calzaghe and Kessler at the top.

Middles OK.

Junior Middle has Ouma and Powell above the field, despite really beating no-one in particular. Most people have DLH deservadly in their p4p top 20, certainly top 30 and yet he lingers at place 8.

I could go on, but it's so omnipresent that it's not really worth it. There are so many examples because of the points I listed in 'position' post. It is exactly as you would expect, if you ignore the domain space.

conan

[mattyp151]

Well, put some "glaring issues" out there.

Hi Matty,

I'm not sure I understand what you mean by glaring issues. I didn't use that term. But if you mean concrete examples of where the rankings are wrong, then I think I've done that already. But to recap.

The long post regarding the light heavy division.

In the heavies, Calvin Brock at 1. Tony Thompson (my god) at 5 after beating everyone's bitch Dom Guinn? Shannon Briggs at 6? Have a look at the fighters that Briggs has been beating. Surface level OKish, but then look at the fighter's they've beaten. Briggs has been brilliant, he has been padding the record with fighters who have obscenely been padding their records.

At cruiser, the most prominent example is Mormeck at 12. That is such a crime. There are also several inflated records in the top 10.

Super Middles has Beyer at 12?? But at least its got Calzaghe and Kessler at the top.

Middles OK.

Junior Middle has Ouma and Powell above the field, despite really beating no-one in particular. Most people have DLH deservadly in their p4p top 20, certainly top 30 and yet he lingers at place 8.

I could go on, but it's so omnipresent that it's not really worth it. There are so many examples because of the points I listed in 'position' post. It is exactly as you would expect, if you ignore the domain space.

conan

So, because you see it differently than everyone else, and you never once mentioned predictability (which the ENTIRE rating system is based on), I have to question what you accept as a good system. Is it a system that you agree with, is a system that isn't based on predictability? I don't get your argument since you don't seem to be basing your opinion on what the system actually is, but rather are just saying what you think it should be.

You ask 10 people to rank the top 10 LHW, you get 10 different lists. Simple as that. Just because it's not what you would expect and not how you would make a rating system doesn't make it a bad system.

[conan_the_cribber]

Well, put some "glaring issues" out there.

Hi Matty,

I'm not sure I understand what you mean by glaring issues. I didn't use that term. But if you mean concrete examples of where the rankings are wrong, then I think I've done that already. But to recap.

The long post regarding the light heavy division.

In the heavies, Calvin Brock at 1. Tony Thompson (my god) at 5 after beating everyone's bitch Dom Guinn? Shannon Briggs at 6? Have a look at the fighters that Briggs has been beating. Surface level OKish, but then look at the fighter's they've beaten. Briggs has been brilliant, he has been padding the record with fighters who have obscenely been padding their records.

At cruiser, the most prominent example is Mormeck at 12. That is such a crime. There are also several inflated records in the top 10.

Super Middles has Beyer at 12?? But at least its got Calzaghe and Kessler at the top.

Middles OK.

Junior Middle has Ouma and Powell above the field, despite really beating no-one in particular. Most people have DLH deservadly in their p4p top 20, certainly top 30 and yet he lingers at place 8.

I could go on, but it's so omnipresent that it's not really worth it. There are so many examples because of the points I listed in 'position' post. It is exactly as you would expect, if you ignore the domain space.

conan

So, because you see it differently than everyone else, and you never once mentioned predictability (which the ENTIRE rating system is based on), I have to question what you accept as a good system. Is it a system that you agree with, is a system that isn't based on predictability? I don't get your argument since you don't seem to be basing your opinion on what the system actually is, but rather are just saying what you think it should be.

You ask 10 people to rank the top 10 LHW, you get 10 different lists. Simple as that. Just because it's not what you would expect and not how you would make a rating system doesn't make it a bad system.

I give up, it is pointless for us to continue Matty. In at least seven posts I mention predictability. And now you claim I dont mention predictability.

You ask me for some "glaring issues". I post I dont know what you mean, but answer with a dozen problems in the rankings. Instead of acknowledging even one of these issues, you change the topic again and say, you dont know what I want. I stated very, very clearly what my stance is, 14 points in all but you are unwilling to pick even one of them and discuss it. Instead you get all mixed up when I bring in Coke vs Pepsi and Bowe, and when I point this out, you get all offended.

It is a bottomless pit discussing this with you. Sorry I tried. I picked your lines apart one by one even. But you are not up to it.

Bring on computerrank. At least he understands what I'm talking about.

conan

[conan_the_cribber]

Agggh! It pains me to say this but I agree with one of Conan's points:

4) The statements about "great predictability" are misfounded. They have no root in mathematical theory e.g. 10000088 might be the optimum for home town advantage.

I cant get my head around this either. All the values for a win are a multiple of 5 and this does seem strange.

It is possible that 2 completelydifferent systems could give the same measure of predictability but each give a completely different ratings list with one "looking" (I hate that phrase but let's go with it) better than the other.

I have to hold my hands up I've tried many different systems for MANCRANK and one of them was if it looked wrong or not.

However, I still like the idea of a completely unbiased system which does not take into account personal views. Since the IBO ones don't say how they calculate them I have to respect Boxrec's more particularly as they are striving to improve and do look a lot better to me than they used to (particularly the all-time ratings) I like the fact that it gives a statistical comparison for P4P and lets you compare fighters from different eras (like ELO ratings)

There's a lot of things I like about the ratings but I do admit that Conan has a couple of good points although the way he puts them over can be annoying though humorous at times too. Elephants indeed. Did you know there are footy rankings called "elephant Rankings" !?

Ahhhh there is hope....

No I didn't know that there are footy rankings called elephant rankings. I'll look them up.

I'll try to be less annoying and more humorous. But I have no more time left to waste on people who can't follow an argument. It's just not worth writing an A4 page full of an argument to get just two lines of response, about a topic that I didn't even mention.

conan

[conan_the_cribber]

Hi,

here's a mathematical puzzle for you all. Especially for computerrank (Martin), he'll enjoy this one.

How do you optimize the predictability factor?

It's simple. What you do is you promote the guys who have inflated records way up to the top of the pile. This is good because for every crossroads fight they fight, they will fight 20 stiffs. This is great for the predictability, because that means the system says 20 wins and maybe only one loss. Predictability of 95%

The higher you promote these record padders, the less likely they are to fight someone above or near them. Because their careers are carefully managed, they continue to profit from the predicability cycle.

What's really good, is when a record padder meets another record padder as in Shannon Briggs vs Koval. Briggs gets a major boost by beating someone, that should never have been so high. And bingo, you have a place in the top 10.

Recognise anything anybody? The goal for a high predictability is a vicious cycle. Fighters who take no risks get promoted, because their fights are so predictable. The predictability of their victory against a hand picked opponent, increases the overall predictability statistic.

That's why there are so many record padders in the list. Not because they do well when they get there, it's because they fight predictably weak opponents and this increases the predicability stat.

What do you think Martin? Recognise the cyclical problem?

conan

[JCS]

I'd like to also point out that the following can have adverse effects on any computerized system:

1. Bad judge/ref scoring. We drill down pretty far on the judge's score cards. If they score a fight too wide, too close, or incorrectly, it will adversely affect ratings.

2. Inaccurate Records. Obviously, this can cause variations.

3. Omitted Information. Missing fights are bad, as well as missing scores. A generic split decision w/ no scores will change ratings more than your razor thin split decision.

4. Lengthy Inactivity Reasons. How can you differentiate between Mesi's inactivity and De La Hoya's inactivity? Mesi obviously lost 2 years.. while Oscar has obviously kept in shape, etc.

[JCS]

What about Conan's "puzzle" tho?

I was saving it for Martin..

[mattyp151]

Hi,

here's a mathematical puzzle for you all. Especially for computerrank (Martin), he'll enjoy this one.

How do you optimize the predictability factor?

It's simple. What you do is you promote the guys who have inflated records way up to the top of the pile. This is good because for every crossroads fight they fight, they will fight 20 stiffs. This is great for the predictability, because that means the system says 20 wins and maybe only one loss. Predictability of 95%

The higher you promote these record padders, the less likely they are to fight someone above or near them. Because their careers are carefully managed, they continue to profit from the predicability cycle.

What's really good, is when a record padder meets another record padder as in Shannon Briggs vs Koval. Briggs gets a major boost by beating someone, that should never have been so high. And bingo, you have a place in the top 10.

Recognise anything anybody? The goal for a high predictability is a vicious cycle. Fighters who take no risks get promoted, because their fights are so predictable. The predictability of their victory against a hand picked opponent, increases the overall predictability statistic.

That's why there are so many record padders in the list. Not because they do well when they get there, it's because they fight predictably weak opponents and this increases the predicability stat.

What do you think Martin? Recognise the cyclical problem?

conan

I thought JCS mentioned there was a padded record factor, called the "Goodson Factor" instituted to push guys who beat cupcakes back. The king of the padded records is going to have to rise because he has a pretty record, and has the best opposition out of the crew. Briggs is a KO guy, KO's score big points, and he has beaten guys with very pretty records all by fast KO. Not the best resume, but he's got a title fight with Wlad in Nov...so it obviously worked, and the system says its not unwinnable for Briggs.

[conan_the_cribber]

I'd like to also point out that the following can have adverse effects on any computerized system:

1. Bad judge/ref scoring. We drill down pretty far on the judge's score cards. If they score a fight too wide, too close, or incorrectly, it will adversely affect ratings.

2. Inaccurate Records. Obviously, this can cause variations.

3. Omitted Information. Missing fights are bad, as well as missing scores. A generic split decision w/ no scores will change ratings more than your razor thin split decision.

4. Lengthy Inactivity Reasons. How can you differentiate between Mesi's inactivity and De La Hoya's inactivity? Mesi obviously lost 2 years.. while Oscar has obviously kept in shape, etc.

Hi JCS,

All of your points are valid. You cant do anything about 1 or 2 or 3, you can only make the best records with the data you have. For these factors, a computer algoirthm is actually at a disadvantage to a human. A human can factor in things like Vitali beating the snot out of Byrd for 9 rounds before having to pull out injured. Point 4, inactivity should be part of any decent algorithm, but there should be no difference for the reason.

conan

[mattyp151]

I'd like to also point out that the following can have adverse effects on any computerized system:

1. Bad judge/ref scoring. We drill down pretty far on the judge's score cards. If they score a fight too wide, too close, or incorrectly, it will adversely affect ratings.

2. Inaccurate Records. Obviously, this can cause variations.

3. Omitted Information. Missing fights are bad, as well as missing scores. A generic split decision w/ no scores will change ratings more than your razor thin split decision.

4. Lengthy Inactivity Reasons. How can you differentiate between Mesi's inactivity and De La Hoya's inactivity? Mesi obviously lost 2 years.. while Oscar has obviously kept in shape, etc.

Hi JCS,

All of your points are valid. You cant do anything about 1 or 2 or 3, you can only make the best records with the data you have. For these factors, a computer algoirthm is actually at a disadvantage to a human. A human can factor in things like Vitali beating the snot out of Byrd for 9 rounds before having to pull out injured. Point 4, inactivity should be part of any decent algorithm, but there should be no difference for the reason.

conan

But, the reason behind inactivity matters. A guy who got KOed and wants to go back to the drawing board still trains while he is inactive, such as Jaidon Codrington. However, a guy like Mesi, who had a legal battle, probably wasn't training much over the 2 years he was absent, and seems to look worse fight to fight. I would venture to say Ike Quartey wasn't training much over his 5 year absence, and he looks better with every fight. Reason for inactivity matters big time, but you'd have to go case by case, and that would take forever.

I'm sure if Phil N'Dou came back, he'd be in the same boat as Mesi, and not Codrington or Quartey.

[conan_the_cribber]

I'd like to also point out that the following can have adverse effects on any computerized system:

1. Bad judge/ref scoring. We drill down pretty far on the judge's score cards. If they score a fight too wide, too close, or incorrectly, it will adversely affect ratings.

2. Inaccurate Records. Obviously, this can cause variations.

3. Omitted Information. Missing fights are bad, as well as missing scores. A generic split decision w/ no scores will change ratings more than your razor thin split decision.

4. Lengthy Inactivity Reasons. How can you differentiate between Mesi's inactivity and De La Hoya's inactivity? Mesi obviously lost 2 years.. while Oscar has obviously kept in shape, etc.

Hi JCS,

All of your points are valid. You cant do anything about 1 or 2 or 3, you can only make the best records with the data you have. For these factors, a computer algoirthm is actually at a disadvantage to a human. A human can factor in things like Vitali beating the snot out of Byrd for 9 rounds before having to pull out injured. Point 4, inactivity should be part of any decent algorithm, but there should be no difference for the reason.

conan

But, the reason behind inactivity matters. A guy who got KOed and wants to go back to the drawing board still trains while he is inactive, such as Jaidon Codrington. However, a guy like Mesi, who had a legal battle, probably wasn't training much over the 2 years he was absent, and seems to look worse fight to fight. I would venture to say Ike Quartey wasn't training much over his 5 year absence, and he looks better with every fight. Reason for inactivity matters big time, but you'd have to go case by case, and that would take forever.

I'm sure if Phil N'Dou came back, he'd be in the same boat as Mesi, and not Codrington or Quartey.

Correct Matty. It does depend on what you do while you are out and sometimes you are out injured and can do nothing. Still you might be smart like DLH is or Sugar Ray Leonard was and train really hard whilst you are off. But because you cant factor it in, you just have to ignore it.

conan

[conan_the_cribber]

Seeming as Martin is not here to answer the question about the vicious cycle or predictability, lets try something else. Lets assume I'm new to boxrec. 1st question,

"What's the difference between ratings and rankings?".

thanks

newbie conan

[mattyp151]

Hi JCS,

All of your points are valid. You cant do anything about 1 or 2 or 3, you can only make the best records with the data you have. For these factors, a computer algoirthm is actually at a disadvantage to a human. A human can factor in things like Vitali beating the snot out of Byrd for 9 rounds before having to pull out injured. Point 4, inactivity should be part of any decent algorithm, but there should be no difference for the reason.

conan

But, the reason behind inactivity matters. A guy who got KOed and wants to go back to the drawing board still trains while he is inactive, such as Jaidon Codrington. However, a guy like Mesi, who had a legal battle, probably wasn't training much over the 2 years he was absent, and seems to look worse fight to fight. I would venture to say Ike Quartey wasn't training much over his 5 year absence, and he looks better with every fight. Reason for inactivity matters big time, but you'd have to go case by case, and that would take forever.

I'm sure if Phil N'Dou came back, he'd be in the same boat as Mesi, and not Codrington or Quartey.

Correct Matty. It does depend on what you do while you are out and sometimes you are out injured and can do nothing. Still you might be smart like DLH is or Sugar Ray Leonard was and train really hard whilst you are off. But because you cant factor it in, you just have to ignore it.

conan

But you don't ignore while picking fights. If Mesi were to take on a top 15 heavyweight in 2004, I may have leaned his way, since his injury, not a frigging chance. There is no way for a system to calculate that, so it's not going to factor it in, and thus be wrong. You can't personify a machine with just numbers. I think you're expecting too much.

[JCS]

But, the reason behind inactivity matters. A guy who got KOed and wants to go back to the drawing board still trains while he is inactive, such as Jaidon Codrington. However, a guy like Mesi, who had a legal battle, probably wasn't training much over the 2 years he was absent, and seems to look worse fight to fight. I would venture to say Ike Quartey wasn't training much over his 5 year absence, and he looks better with every fight. Reason for inactivity matters big time, but you'd have to go case by case, and that would take forever.

I'm sure if Phil N'Dou came back, he'd be in the same boat as Mesi, and not Codrington or Quartey.

Correct Matty. It does depend on what you do while you are out and sometimes you are out injured and can do nothing. Still you might be smart like DLH is or Sugar Ray Leonard was and train really hard whilst you are off. But because you cant factor it in, you just have to ignore it.

conan

But you don't ignore while picking fights. If Mesi were to take on a top 15 heavyweight in 2004, I may have leaned his way, since his injury, not a frigging chance. There is no way for a system to calculate that, so it's not going to factor it in, and thus be wrong. You can't personify a machine with just numbers. I think you're expecting too much.

BTW, Mesi is now down to #22, from #13. Progress baby, progress.

[conan_the_cribber]

I'd say the rating was the numerical value ie 1724 While the ranking places the ratings in descending order where first is equivalent to the highest rating.

Something like that I'd guess.

That's what I'd say too. Tiger Woods is the number #1 ranked golfer because he has a rating of 16.88 or Federer is the number one ranked tennis player because he has 6200pts or some chess dude is number #1 because he has 2850 pts.

That's why I can't understand why people claim this is not a ranking system. First of all people are rated, then they are ranked according to the rating.

Anybody disagree?

conan

[conan_the_cribber]

BTW, Mesi is now down to #22, from #13. Progress baby, progress.

JCS,

why do you view this as progress? If predicability is your goal and these are not rankings, then surely you dont care if Mesi comes in at #5, #10, #15 or #22, as long as it contributes to the overall predictability statistic.

Why is Mesi at 13 wrong and 22 right ?

conan

[computerrank]

Hi,

here's a mathematical puzzle for you all. Especially for computerrank (Martin), he'll enjoy this one.

How do you optimize the predictability factor?

It's simple. What you do is you promote the guys who have inflated records way up to the top of the pile. This is good because for every crossroads fight they fight, they will fight 20 stiffs. This is great for the predictability, because that means the system says 20 wins and maybe only one loss. Predictability of 95%

The higher you promote these record padders, the less likely they are to fight someone above or near them. Because their careers are carefully managed, they continue to profit from the predicability cycle.

What's really good, is when a record padder meets another record padder as in Shannon Briggs vs Koval. Briggs gets a major boost by beating someone, that should never have been so high. And bingo, you have a place in the top 10.

Recognise anything anybody? The goal for a high predictability is a vicious cycle. Fighters who take no risks get promoted, because their fights are so predictable. The predictability of their victory against a hand picked opponent, increases the overall predictability statistic.

That's why there are so many record padders in the list. Not because they do well when they get there, it's because they fight predictably weak opponents and this increases the predicability stat.

What do you think Martin? Recognise the cyclical problem?

conan

Conan,

... I anwered this question in my first answer to your first request some 30 messages ago - now again - the majority of these "overrated" boxers will indeed defeat a boxer at the level of their current rating - as soon as they try - this is not my opinion - I analyzed this - it might look funny and you might not like it, but these are the facts:

Regarding your statement, that boxers never should be in the tops without defeating opponents near the same level:

- We studied this aspect very deeply
- We compared the top boxers' ratings and the level of their opponents' ratings and found (by comparing the pre-bout ratings and the results):
- Even for a boxer, only boxing and defeating opponents 300 and 400 points lower - the probability of
-- winning or losing to an opponent of his own rating level is about 50%,
-- is more than 50% for opponents, rated only a little bit lower
-- and of course less than 50% for opponents, rated only a little bit higher
-- so the ratings are quite consistent in this sense

This simply means, the current ratings are quite correct - and it means that these boxers have quite a good chance to defeat opponents of definite higher level than they might have chosen before - as soon as they step up.

This also means - and I understand this - we all would prefer for sure, to see every boxer box at his approriate level. But unluckily they don't.

And - in spite of this, the majority of these risk-minimizing boxers perform better than most would admit, as soon as they dare.

These ratings are not made to punish boxers - they show the reality of based exspectations.

The traditional ranking style systematically under-estimates the performance of such boxers and so over-estimates the performance of the old guys.

By the way - ELO ratings are also performance ratings - and performance ratings are regarded as the standard of representing a rating by statistical, mathematical means.

Today, we have a little bit more to consider, than only archaicly rely on having sent an opponent to the floor prior to promoting him in the ranks ... altough most would still prefer this method, I fear ...

[conan_the_cribber]

Hi,

here's a mathematical puzzle for you all. Especially for computerrank (Martin), he'll enjoy this one.

How do you optimize the predictability factor?

It's simple. What you do is you promote the guys who have inflated records way up to the top of the pile. This is good because for every crossroads fight they fight, they will fight 20 stiffs. This is great for the predictability, because that means the system says 20 wins and maybe only one loss. Predictability of 95%

The higher you promote these record padders, the less likely they are to fight someone above or near them. Because their careers are carefully managed, they continue to profit from the predicability cycle.

What's really good, is when a record padder meets another record padder as in Shannon Briggs vs Koval. Briggs gets a major boost by beating someone, that should never have been so high. And bingo, you have a place in the top 10.

Recognise anything anybody? The goal for a high predictability is a vicious cycle. Fighters who take no risks get promoted, because their fights are so predictable. The predictability of their victory against a hand picked opponent, increases the overall predictability statistic.

That's why there are so many record padders in the list. Not because they do well when they get there, it's because they fight predictably weak opponents and this increases the predicability stat.

What do you think Martin? Recognise the cyclical problem?

conan

Conan,

... I anwered this question in my first answer to your first request some 30 messages ago - now again - the majority of these "overrated" boxers will indeed defeat a boxer at the level of their current rating - as soon as they try - this is not my opinion - I analyzed this - it might look funny and you might not like it, but these are the facts:

Regarding your statement, that boxers never should be in the tops without defeating opponents near the same level:

- We studied this aspect very deeply
- We compared the top boxers' ratings and the level of their opponents' ratings and found (by comparing the pre-bout ratings and the results):
- Even for a boxer, only boxing and defeating opponents 300 and 400 points lower - the probability of
-- winning or losing to an opponent of his own rating level is about 50%,
-- is more than 50% for opponents, rated only a little bit lower
-- and of course less than 50% for opponents, rated only a little bit higher
-- so the ratings are quite consistent in this sense

This simply means, the current ratings are quite correct - and it means that these boxers have quite a good chance to defeat opponents of definite higher level than they might have chosen before - as soon as they step up.

This also means - and I understand this - we all would prefer for sure, to see every boxer box at his approriate level. But unluckily they don't.

And - in spite of this, the majority of these risk-minimizing boxers perform better than most would admit, as soon as they dare.

These ratings are not made to punish boxers - they show the reality of based exspectations.

The traditional ranking style systematically under-estimates the performance of such boxers and so over-estimates the performance of the old guys.

By the way - ELO ratings are also performance ratings - and performance ratings are regarded as the standard of representing a rating by statistical, mathematical means.

Today, we have a little bit more to consider, than only archaicly rely on having sent an opponent to the floor prior to promoting him in the ranks ... altough most would still prefer this method, I fear ...

Yawwwwnnnnnnnnn. I thought you were better than that Martin.

You come out after 30 posts and just cut and paste the last respone!? Shall I come out now and cut and paste all my posts so that the questions remain the same.

I present you a completely logical argument as to why the record padders are so high. I debug your myths about being the 'best' or even 'optimal' predicative rankings. I present you devistating feedback about the state of the light heavys based on real life events. Nobody supports the Dawson Stipes theory. Your peers vote that the rankings suck. None of your followers can argue why this is not a ranking system.

And all you say to this is nothing. How embarassing.

As you are unable or unwilling to argue any of my 14 points, or even to acknowledge the rather obvious vicious cycle problem caused by striving solely for high predictability, I take it that you concede my points.

conan

[computerrank]

@conan

... as you cannot detect the answer to your predictablity question in my post, we will not come together ... I suggest, use your time ... it will do with mine ...

[mattyp151]

Hi,

here's a mathematical puzzle for you all. Especially for computerrank (Martin), he'll enjoy this one.

How do you optimize the predictability factor?

It's simple. What you do is you promote the guys who have inflated records way up to the top of the pile. This is good because for every crossroads fight they fight, they will fight 20 stiffs. This is great for the predictability, because that means the system says 20 wins and maybe only one loss. Predictability of 95%

The higher you promote these record padders, the less likely they are to fight someone above or near them. Because their careers are carefully managed, they continue to profit from the predicability cycle.

What's really good, is when a record padder meets another record padder as in Shannon Briggs vs Koval. Briggs gets a major boost by beating someone, that should never have been so high. And bingo, you have a place in the top 10.

Recognise anything anybody? The goal for a high predictability is a vicious cycle. Fighters who take no risks get promoted, because their fights are so predictable. The predictability of their victory against a hand picked opponent, increases the overall predictability statistic.

That's why there are so many record padders in the list. Not because they do well when they get there, it's because they fight predictably weak opponents and this increases the predicability stat.

What do you think Martin? Recognise the cyclical problem?

conan

Conan,

... I anwered this question in my first answer to your first request some 30 messages ago - now again - the majority of these "overrated" boxers will indeed defeat a boxer at the level of their current rating - as soon as they try - this is not my opinion - I analyzed this - it might look funny and you might not like it, but these are the facts:

Regarding your statement, that boxers never should be in the tops without defeating opponents near the same level:

- We studied this aspect very deeply
- We compared the top boxers' ratings and the level of their opponents' ratings and found (by comparing the pre-bout ratings and the results):
- Even for a boxer, only boxing and defeating opponents 300 and 400 points lower - the probability of
-- winning or losing to an opponent of his own rating level is about 50%,
-- is more than 50% for opponents, rated only a little bit lower
-- and of course less than 50% for opponents, rated only a little bit higher
-- so the ratings are quite consistent in this sense

This simply means, the current ratings are quite correct - and it means that these boxers have quite a good chance to defeat opponents of definite higher level than they might have chosen before - as soon as they step up.

This also means - and I understand this - we all would prefer for sure, to see every boxer box at his approriate level. But unluckily they don't.

And - in spite of this, the majority of these risk-minimizing boxers perform better than most would admit, as soon as they dare.

These ratings are not made to punish boxers - they show the reality of based exspectations.

The traditional ranking style systematically under-estimates the performance of such boxers and so over-estimates the performance of the old guys.

By the way - ELO ratings are also performance ratings - and performance ratings are regarded as the standard of representing a rating by statistical, mathematical means.

Today, we have a little bit more to consider, than only archaicly rely on having sent an opponent to the floor prior to promoting him in the ranks ... altough most would still prefer this method, I fear ...

Yawwwwnnnnnnnnn. I thought you were better than that Martin.

You come out after 30 posts and just cut and paste the last respone!? Shall I come out now and cut and paste all my posts so that the questions remain the same.

I present you a completely logical argument as to why the record padders are so high. I debug your myths about being the 'best' or even 'optimal' predicative rankings. I present you devistating feedback about the state of the light heavys based on real life events. Nobody supports the Dawson Stipes theory. Your peers vote that the rankings suck. None of your followers can argue why this is not a ranking system.

And all you say to this is nothing. How embarassing.

As you are unable or unwilling to argue any of my 14 points, or even to acknowledge the rather obvious vicious cycle problem caused by striving solely for high predictability, I take it that you concede my points.

conan

Here ya go Conan, a suggestion.

Get off your fvckin high horse, stop bitching about ranking/rating whatever, and stop coming to the site if you dislike the system so much. You come around once in a blue moon, and you want to change the system because you think it's bad....shit doesn't work that way.

If you want to have a ranking system that you agree with, make your own site and do it yourself. You've had enough time to sit and bitch about this system, so that Bullshit excuse "I don't have the time" is moot.

You've gone way past being constructive and blown right into being arrogant.

[conan_the_cribber]

Here ya go Conan, a suggestion.

Get off your fvckin high horse, stop bitching about ranking/rating whatever, and stop coming to the site if you dislike the system so much. You come around once in a blue moon, and you want to change the system because you think it's bad....shit doesn't work that way.

If you want to have a ranking system that you agree with, make your own site and do it yourself. You've had enough time to sit and bitch about this system, so that Bullshit excuse "I don't have the time" is moot.

You've gone way past being constructive and blown right into being arrogant.

For once Matty, you're right.

It is probably quicker to code up alternative rankings than convince Martin or his fans that even one fraction of the algorithm is wrong.

Some info for you Matty. I actually chose a very arrogant persona this time, in order to provoke debate. Last time Martin and I conversed, it was very cordial and co-operative. If I'm not mistaken, I even suggested the home town advantage and amongst others another factor, wear and tear on the fighter. It was ironic that the home town advantge got slapped back in my face. However, cheeriness and cordiality got me got me absolutely nowhere. I thought I'd try another tack, but Martin is, if nothing else, very convinced in his ways. I'm surprised that he's able to convince others, but I guess that his work ethic impresses. Judging by many other posts on the site, not so many people are convinced with the results of his work. If you are prepared to accept Tony Thompson as a top 6 fighter, then jolly good luck to you.

@Matty. I was particularly hard on you Matty. For that I apologise. You are probably a decent bloke and probably excellent company down at the pub, enjoying a beer and watching a fight. I don't think we should design software together, but hey, I dont go the pub with my boss either. Maybe I will go away, maybe I will stay, but I promise to be more respectful to you in the future.

@Martin. Of course I read your answer to my question in your post. The answer is predictability, "predictability". Boxers form a pyramid. There are very few elite fighters at the top, a few champion at the class, more contendors, even more journeymen etc etc. The top of the pyramid is narrow, and that sometimes it's occupied by 2 or 3 fighters for 5 years at a time. As such it's only a natural conclusion that most people dont make it that high. There are a lot more people with 20-0 records that dont go on to become a champ, than there are people with 20-0 records that do. Given that, I'm surprised with your claim that when push comes to shove, a slight majority of up and comers or record padders do win their crossroads fights against established fighters. It is very counter-intuitive, there are a lot more Hector Camacho Jr's and Rydell Bookers than Artur Abrahams. But I assume your a man of integrity and that you can prove your claim. I hope you make a fortune at the bookies with all this predictive stuff, cos your predictions are often way out of line with the bookmakers odds. Good luck to you, you stubborn mule.

@CopyWebCat. Thanks for some support, at least you had an open mind to some of my points.

@JCS. Before I go (from this thread at least), I would appreciate very much if you answered one question. Quite often I asked you something and you didn't answer, but I would like you to answer just one question. Please. It was the last one I posed. Why is Mesi at 22 "progress" in your eyes? Could it be, that it "looked" better? By the way. That post on the other thread about hearing a little girl crying, was a howler. I was grinning ear to ear after that one. That was damm funny.

See ya round.

conan