Ratings - please read before commenting - Archived

[Daedalus]

It is the natural rating. At start of his career each boxers gets 0.15 hidden prior wins and 2.85 hidden prior losses against an artificial opponent with rating 5 for proboxing men and 50 for proboxing women. These prior results tie each boxer to an initial start point.

There are no negative values. The values are somewhere between 0.0001 and 9999.

Ok, so if 1-pt boxers keep losing, they go below 1 but the display stays 1.000. Is that correct, or are more hidden wins added to bring up to 1?
I know past formulas displayed down to 0, so why the 1-pt floor now?

There is no 1 point floor, neither for the calulation nor for the display. What example do you look at?

Sorry can't find it anymore. I must've been looking at a page with all 1.000 and thought it was the last.

[Offthehook]

All Whole-History Ratings of all boxers change with each new bout as each new bout changes the likelihood of all other results.

you should be a politician with that answer. So basically the more maxi hughes wins, the higher his previous opponents (win or lose) go?

So, where is it?

Nobody was interested in that - it is off.

This is a politician's answer, instead of "we can't do it"

i think i speak on behalf of every knowledgable boxing fan when i say, the rankings are a bit bullshit. Youve got africans ranked higher than guys who have challenged for world or euro titles. Uve rewardede certain fighters with lots of points for beating either unproven or weak opposition, and others whove fought at a higher level less.

Mybe boxrec would be better served as a matchmakers tool and a forum, rather than trying to rank fighters based on an algorithm that seems to be subjective.

[computerrank]

i think i speak on behalf of every knowledgable boxing fan when i say, the rankings are a bit bullshit. Youve got africans ranked higher than guys who have challenged for world or euro titles. Uve rewardede certain fighters with lots of points for beating either unproven or weak opposition, and others whove fought at a higher level less.
...

Boxrec ratings do not mind titles. And if you inspect, who is ranked by the organisations and who gets the title shots, then it is obvious, why Boxrec ratings do not mind. In real world getting a shot for titles doesn't correlate with proven achievements,

[JCS]

Mybe boxrec would be better served as a matchmakers tool and a forum, rather than trying to rank fighters based on an algorithm that seems to be subjective.

Isn't every system going to be at least somewhat subjective? After all, the computer can't pick the goal outcome..

[John]

It isn't subjective, the metric is solely prediction accuracy not human opinion.

[JCS]

It isn't subjective, the metric is solely prediction accuracy not human opinion.

If Martin gave a boost to boxers with a larger reach, would you be OK with factoring that in? What if a 0.0005 higher prediction rate was possible, but the ratings would take a month to compute?

There are some levels of constraint, common sense and other things that a human must apply. I suppose you could refer to this as subjectivity... as well as deciding on prediction rate as the primary driver, instead of aligning with an "average" of the expert rankings out there.

[felinoboxing]

Just tell everyone in boxrec that with the P4P Annual rating, they have made a man happy, from a small town in Venezuela. Great job.

[felinoboxing]

To make the annual boxrec rating I see that they took the score at the end of each year.
So in 1967 Ali is on top with 584 points. The thing is, Ali was suspended in April of that year. He was inactive.

[computerrank]

To make the annual boxrec rating I see that they took the score at the end of each year.
So in 1967 Ali is on top with 584 points. The thing is, Ali was suspended in April of that year. I was inactive.

A boxer's annual rating is taken from his last annual bout. So if he was suspended later, it doesn't matter.

[felinoboxing]

It is quite possible that there are boxers in the past who were # 1 in boxrec's P4P but not at the end of some year. These boxers will not be seen in the 1 of these annual rankings. To see the changes in number 1, I suggest a linear P4P # 1. Something like this:
https://boxing.fandom.com/wiki/List_of_ ... _champions

Only numbers 1!

[Offthehook]

i think i speak on behalf of every knowledgable boxing fan when i say, the rankings are a bit bullshit. Youve got africans ranked higher than guys who have challenged for world or euro titles. Uve rewardede certain fighters with lots of points for beating either unproven or weak opposition, and others whove fought at a higher level less.
...

Boxrec ratings do not mind titles. And if you inspect, who is ranked by the organisations and who gets the title shots, then it is obvious, why Boxrec ratings do not mind. In real world getting a shot for titles doesn't correlate with proven achievements,

its not ''boxrec'' that ranks fighters though is it?...you are solely responsible for the rankings process, are you not? Here is the point i have a problem with...

fighter A fights at a lower level than fighter B. Fighter A's 2 best wins are below fighter B's best win. Fighter B also has fought proven men at a higher level and has a draw at true 'world level'.

Fighter A is ranked higher....in fact has double the points.....when confronted computerank doesnt seem to give a direct answer, and talks about 'history' (that is wiped clean) but if the rankings are about the here and now, surely Fighter B should be ranked higher.

In other words i feel that no matter that all the evidence points to fighter B should be ranked higher, things wont change.

If you would like me to assign names to fighter A and B, so people can check the records, id be more than happy to.

Lastly im not having a go computerank, im just merely stating facts, and i may not have as much experience with algorithms as yourself (although i understand how they work) i know fighters, and where essentially they should rank, which is why myself, and many have a problem with the rankings.

[Offthehook]

Mybe boxrec would be better served as a matchmakers tool and a forum, rather than trying to rank fighters based on an algorithm that seems to be subjective.

Isn't every system going to be at least somewhat subjective? After all, the computer can't pick the goal outcome..

that wasnt my point, but surely any computer algorithm should rank fighters who have had better wins, and fought at a higher level above those that have not. If it cant do that, with complete transparency, then it shouldnt bother with a ranking system.

[JCS]

Mybe boxrec would be better served as a matchmakers tool and a forum, rather than trying to rank fighters based on an algorithm that seems to be subjective.

Isn't every system going to be at least somewhat subjective? After all, the computer can't pick the goal outcome..

that wasnt my point, but surely any computer algorithm should rank fighters who have had better wins, and fought at a higher level above those that have not. If it cant do that, with complete transparency, then it shouldnt bother with a ranking system.

Aren't terms like "better wins" and "higher level" subjective, though?

[Manrae]

you should be a politician with that answer. So basically the more maxi hughes wins, the higher his previous opponents (win or lose) go?

Nobody was interested in that - it is off.

This is a politician's answer, instead of "we can't do it"

i think i speak on behalf of every knowledgable boxing fan when i say, the rankings are a bit bullshit. Youve got africans ranked higher than guys who have challenged for world or euro titles. Uve rewardede certain fighters with lots of points for beating either unproven or weak opposition, and others whove fought at a higher level less.

Mybe boxrec would be better served as a matchmakers tool and a forum, rather than trying to rank fighters based on an algorithm that seems to be subjective.

Speak for yourself, racist.

[JCS]

This is a politician's answer, instead of "we can't do it"

i think i speak on behalf of every knowledgable boxing fan when i say, the rankings are a bit bullshit. Youve got africans ranked higher than guys who have challenged for world or euro titles. Uve rewardede certain fighters with lots of points for beating either unproven or weak opposition, and others whove fought at a higher level less.

Mybe boxrec would be better served as a matchmakers tool and a forum, rather than trying to rank fighters based on an algorithm that seems to be subjective.

Speak for yourself, racist.

He's obviously speaking about the African boxing scene being less prestigious than that in Europe and other places in the world. Don't pull the racist accusations crap..

[Offthehook]

This is a politician's answer, instead of "we can't do it"

i think i speak on behalf of every knowledgable boxing fan when i say, the rankings are a bit bullshit. Youve got africans ranked higher than guys who have challenged for world or euro titles. Uve rewardede certain fighters with lots of points for beating either unproven or weak opposition, and others whove fought at a higher level less.

Mybe boxrec would be better served as a matchmakers tool and a forum, rather than trying to rank fighters based on an algorithm that seems to be subjective.

Speak for yourself, racist.

lol how is that a racist comment?....educate yourself. African boxing is at a lower level than european, thats factual. some of the guys ranked high in africa, have fought no one....people like

https://boxrec.com/en/proboxer/697487

https://boxrec.com/en/proboxer/613045

ranked above ex world title contenders and proven european class fighters. look at their resumes....then shut the fck up!

people who shout racist usually are those deflecting attention from themselves. My gf is half ghanian......like i say educate yourself....dickhead!

[Offthehook]

i think i speak on behalf of every knowledgable boxing fan when i say, the rankings are a bit bullshit. Youve got africans ranked higher than guys who have challenged for world or euro titles. Uve rewardede certain fighters with lots of points for beating either unproven or weak opposition, and others whove fought at a higher level less.

Mybe boxrec would be better served as a matchmakers tool and a forum, rather than trying to rank fighters based on an algorithm that seems to be subjective.

Speak for yourself, racist.

He's obviously speaking about the African boxing scene being less prestigious than that in Europe and other places in the world. Don't pull the racist accusations crap..

thank you JCS...thought i was speaking plainly enough! Absolutely disgusting comment from that half wit!

[JCS]

When it comes to these smaller scenes, the real sh1t happens when these guys finally branch out and fight boxers in bigger populations but....

- Are they getting put in shitty situations? Short notice? Have to go up a division?
- Are they past their prime when they do it?
- Do they get jobbed on the scorecards?

Is the computer better or worse at rating these unknowns? Most of the time, we'll never really know.

[Daedalus]

When it comes to these smaller scenes, the real sh1t happens when these guys finally branch out and fight boxers in bigger populations but....

- Are they getting put in shitty situations? Short notice? Have to go up a division?
- Are they past their prime when they do it?
- Do they get jobbed on the scorecards?

Is the computer better or worse at rating these unknowns? Most of the time, we'll never really know.

WHR really likes Jack Carroll, who never fought outside Australia or met anyone with over 200 pts, yet it placed him in the top 20 all time pound-for-pound.

[JCS]

When it comes to these smaller scenes, the real sh1t happens when these guys finally branch out and fight boxers in bigger populations but....

- Are they getting put in shitty situations? Short notice? Have to go up a division?
- Are they past their prime when they do it?
- Do they get jobbed on the scorecards?

Is the computer better or worse at rating these unknowns? Most of the time, we'll never really know.

WHR really likes Jack Carroll, who never fought outside Australia or met anyone with over 200 pts, yet it placed him in the top 20 all time pound-for-pound.

Maybe we can find an expert of the 1920s, 1930s Australian boxing scene.. must've been very prolific.

[computerrank]

Saturday, October 24th
Sergey Lipinets vs. Custio Clayton
"5-star" fighter vs. "4-star" fighter
Listed as a "no-star" fight.

Maybe it happens, when the commission approves the bout ...
I now again instant updated it.

[computerrank]

TV Coverage starts within the hour:

Roman Gonzalez ("5-star" fighter) vs. Israel Gonzalez ("3½ star" fighter) listed as "no-star" bout.
Julio Cesar Martinez ("4½ star" fighter) vs. Moises Calleros ("2½ star" fighter) listed as a "no-star" bout.

Still not clear, why it happens.

[jujigatame]

Can you shed any light onto the rating of Subriel Matias? The WHR progression shows he's been hovering between 25-55 for the last few years and had 38 points for his last fight in February, but if you look at his rating right now he only has 11 points? What could have caused such a huge decline?

[computerrank]

Can you shed any light onto the rating of Subriel Matias? The WHR progression shows he's been hovering between 25-55 for the last few years and had 38 points for his last fight in February, but if you look at his rating right now he only has 11 points? What could have caused such a huge decline?

Here is something wrong

Matias lost against Ananyan in his last bout. Ananyan came from nowhere. Pure WHR still rates Ananyan at 11.67 and Matias at 38.42.

But Boxrec has a rule, that the winner is always above the loser for at least 1 year.

So Ananyan should be a bit above the mean of their ratings, and Matias should be a bit below that. But the margin is far too high

[jujigatame]

Wait, I actually found a bug?

I will wear my honorary BoxRec QA badge with pride.