What's been up with the ratings this year?
What's been up with the ratings this year?
I'm curious what's been going on with the rankings over the last year? They used to be fairly stable, if a bit off. However, now they are all over the map and fighters are leaping up to incredibly high ratings for no real good reason. It seems whenever a top 10 fighter in a given division wins, he is suddenly #1. This really needs to be fixed, I've been hearing lots of criticism regarding this lately.
-
computerrank
- Editor
- Posts: 2492
- Joined: 04 Jan 2003, 18:59
Re: What's been up with the ratings this year?
Please look at:Diego wrote:I'm curious what's been going on with the rankings over the last year? They used to be fairly stable, if a bit off. However, now they are all over the map and fighters are leaping up to incredibly high ratings for no real good reason. It seems whenever a top 10 fighter in a given division wins, he is suddenly #1. This really needs to be fixed, I've been hearing lots of criticism regarding this lately.
http://www.boxrec.com/media/index.php/B ... escription
and
http://www.boxrec.com/forum/viewtopic.php?t=28051
first ...
I've read the description and the topic thread, but it simply looks like rationalization for a flawed system.
Let's take a brief look at the current Heavyweight rankings, for example and let's look at this realistically:
It's preposterous that Valuev should jump over Wladimir just by beating Owen Beck, which the system currently ranks 75th.
Calvin Brock suddenly jumps ahead of everyone by beating the guy currently ranked 40th.
Tye Fields is number 8, despite never beating a legitimate top 20 opponent.
Briggs was around 150 a few months ago is all the sudden #6 despite beating no one of note.
Thompson, while coming off a fine win, is now #5.
Brian Minto is #10???
Skelton is ahead of Danny Williams? What the hell?
Malik Scott is ahead of Lyakhovich, Brewster, Byrd, & Chagaev??? Why? Especially Lyakhovich. Brewster and Byrd lose and suddenly they're not even top 20 fighters, while Skelton can lose and is hardly affected?
Ruiz is 26?
Juan Carlos Gomez is 43?
Larry Donald is 51?
The thing is, yes, ratings are imperfect, computer ratings systems can't always take all of the factors into play, etc., etc. But if this is the best we can do, then perhaps we shouldn't do it at all. It really detracts from the rest of the site and makes users question our legimacy when they arrive at the site and see Calvin Brock as the #1 HW, or Chad Dawson as the #1 LHW, Kassim Oua as the #1 LMW, etc. And when did Cotto become #1 at his weight (he was grossly underrated here not very long ago, but now he's #1 after beating Malignaggi?)?
These inconsistencies are rampant and need to be stopped. I hate when I read an article and find a PR agency claiming that Minto is rated '#10 in the Boxrec world ratings!' or some other rubbish. It makes people think that the system is rigged. Bad ratings are one of the vile epidemics plaguing boxing right now, and it's sad to see that we are propogating such.
Boxrec has been a shining glimmer of hope in the boxing game and it's taken years and years to be taken seriously...but all it takes is for Dan Rafael or another major writer to notice these attrocities, and then a mockery is made of us all.
Instead of defending poor rankings, how about we track down the bugs in the system that is causing such and fix them, instead of poorly trying to justify them? That is the issue here...
Let's take a brief look at the current Heavyweight rankings, for example and let's look at this realistically:
It's preposterous that Valuev should jump over Wladimir just by beating Owen Beck, which the system currently ranks 75th.
Calvin Brock suddenly jumps ahead of everyone by beating the guy currently ranked 40th.
Tye Fields is number 8, despite never beating a legitimate top 20 opponent.
Briggs was around 150 a few months ago is all the sudden #6 despite beating no one of note.
Thompson, while coming off a fine win, is now #5.
Brian Minto is #10???
Skelton is ahead of Danny Williams? What the hell?
Malik Scott is ahead of Lyakhovich, Brewster, Byrd, & Chagaev??? Why? Especially Lyakhovich. Brewster and Byrd lose and suddenly they're not even top 20 fighters, while Skelton can lose and is hardly affected?
Ruiz is 26?
Juan Carlos Gomez is 43?
Larry Donald is 51?
The thing is, yes, ratings are imperfect, computer ratings systems can't always take all of the factors into play, etc., etc. But if this is the best we can do, then perhaps we shouldn't do it at all. It really detracts from the rest of the site and makes users question our legimacy when they arrive at the site and see Calvin Brock as the #1 HW, or Chad Dawson as the #1 LHW, Kassim Oua as the #1 LMW, etc. And when did Cotto become #1 at his weight (he was grossly underrated here not very long ago, but now he's #1 after beating Malignaggi?)?
These inconsistencies are rampant and need to be stopped. I hate when I read an article and find a PR agency claiming that Minto is rated '#10 in the Boxrec world ratings!' or some other rubbish. It makes people think that the system is rigged. Bad ratings are one of the vile epidemics plaguing boxing right now, and it's sad to see that we are propogating such.
Boxrec has been a shining glimmer of hope in the boxing game and it's taken years and years to be taken seriously...but all it takes is for Dan Rafael or another major writer to notice these attrocities, and then a mockery is made of us all.
Instead of defending poor rankings, how about we track down the bugs in the system that is causing such and fix them, instead of poorly trying to justify them? That is the issue here...
-
computerrank
- Editor
- Posts: 2492
- Joined: 04 Jan 2003, 18:59
Diego,
I feel your pain, but I canot help, other than to explain the point.
However a rating is made, there is a simple way to find out, whether it is better or worse than an other one.
If you find, that for Rating A more winners of bouts were rated higher before the bout than for Rating B - then Rating A has demonstrated to be the more reliable.
We want to produce the most reliable ratings we can - and we do. We not only look at the ratings, until we feel well - we evaluate this reliability. We look for ways to improve it. And as soon as we find a way to improve it considerably, we will implement this improvement.
This way has been honoured in the past, and will continue to be honoured in the future. This is the only objective way. No individual feelings, no good-old-days, just facts, facts, facts ...
Best regards
Martin
I feel your pain, but I canot help, other than to explain the point.
However a rating is made, there is a simple way to find out, whether it is better or worse than an other one.
If you find, that for Rating A more winners of bouts were rated higher before the bout than for Rating B - then Rating A has demonstrated to be the more reliable.
We want to produce the most reliable ratings we can - and we do. We not only look at the ratings, until we feel well - we evaluate this reliability. We look for ways to improve it. And as soon as we find a way to improve it considerably, we will implement this improvement.
This way has been honoured in the past, and will continue to be honoured in the future. This is the only objective way. No individual feelings, no good-old-days, just facts, facts, facts ...
Best regards
Martin
>Diego,
>
>I feel your pain, but I canot help, other than to explain the point.
>
>However a rating is made, there is a simple way to find out, whether it >is better or worse than an other one.
>
>If you find, that for Rating A more winners of bouts were rated higher >before the bout than for Rating B - then Rating A has demonstrated to >be the more reliable.
>
How is this a better system, and what is this 'Rating B' that you're speaking of? Ratings are not supposed to be predictors of fights, nor who people (or a system) feel is better, they are supposed to be based on merit and achievements alone, which is as close to being objective as is possible in this sport. For example, I feel that Alexander Povetkin could probably beat a few top 15 fighters at this moment, but he simply does not deserve to be rated there as of yet. A system that attempts to rate 'winners of bouts' is inherently unreliable and subjective.
>We want to produce the most reliable ratings we can - and we do. We >not only look at the ratings, until we feel well - we evaluate this >reliability. We look for ways to improve it. And as soon as we find a way >to improve it considerably, we will implement this improvement.
>
I feel the ratings that were present before this sudden sea change were more consistent and valid.
>This way has been honoured in the past, and will continue to be >honoured in the future. This is the only objective way. No individual >feelings, no good-old-days, just facts, facts, facts ...
>
There were never any 'good-old-days', as ratings have been biased for years. However, this current system is insulting to fighters who have actually accomplished a great deal, yet being overlooked in favor of other fighters who simply have not.
What is objective or fact-based about Malik Scott, Tye Fields & Brian Minto being rated ahead of Lyakhovich, Brewster, Byrd, & Williams??? And how can one justify Matt Skelton being ranked ahead of Danny Williams for the time being? If this is a meritocracy, something went horribly wrong.
If we simply want to be mediocre and present ratings that make no logical sense, then we are doing an absolutely superb job.
>Best regards
>
>Martin[/quote]
>
>I feel your pain, but I canot help, other than to explain the point.
>
>However a rating is made, there is a simple way to find out, whether it >is better or worse than an other one.
>
>If you find, that for Rating A more winners of bouts were rated higher >before the bout than for Rating B - then Rating A has demonstrated to >be the more reliable.
>
How is this a better system, and what is this 'Rating B' that you're speaking of? Ratings are not supposed to be predictors of fights, nor who people (or a system) feel is better, they are supposed to be based on merit and achievements alone, which is as close to being objective as is possible in this sport. For example, I feel that Alexander Povetkin could probably beat a few top 15 fighters at this moment, but he simply does not deserve to be rated there as of yet. A system that attempts to rate 'winners of bouts' is inherently unreliable and subjective.
>We want to produce the most reliable ratings we can - and we do. We >not only look at the ratings, until we feel well - we evaluate this >reliability. We look for ways to improve it. And as soon as we find a way >to improve it considerably, we will implement this improvement.
>
I feel the ratings that were present before this sudden sea change were more consistent and valid.
>This way has been honoured in the past, and will continue to be >honoured in the future. This is the only objective way. No individual >feelings, no good-old-days, just facts, facts, facts ...
>
There were never any 'good-old-days', as ratings have been biased for years. However, this current system is insulting to fighters who have actually accomplished a great deal, yet being overlooked in favor of other fighters who simply have not.
What is objective or fact-based about Malik Scott, Tye Fields & Brian Minto being rated ahead of Lyakhovich, Brewster, Byrd, & Williams??? And how can one justify Matt Skelton being ranked ahead of Danny Williams for the time being? If this is a meritocracy, something went horribly wrong.
If we simply want to be mediocre and present ratings that make no logical sense, then we are doing an absolutely superb job.
>Best regards
>
>Martin[/quote]
[/quote]Diego wrote:>Diego,
>
>I feel your pain, but I canot help, other than to explain the point.
>
>However a rating is made, there is a simple way to find out, whether it >is better or worse than an other one.
>
>If you find, that for Rating A more winners of bouts were rated higher >before the bout than for Rating B - then Rating A has demonstrated to >be the more reliable.
>
How is this a better system, and what is this 'Rating B' that you're speaking of? Ratings are not supposed to be predictors of fights, nor who people (or a system) feel is better, they are supposed to be based on merit and achievements alone, which is as close to being objective as is possible in this sport. For example, I feel that Alexander Povetkin could probably beat a few top 15 fighters at this moment, but he simply does not deserve to be rated there as of yet. A system that attempts to rate 'winners of bouts' is inherently unreliable and subjective.
>We want to produce the most reliable ratings we can - and we do. We >not only look at the ratings, until we feel well - we evaluate this >reliability. We look for ways to improve it. And as soon as we find a way >to improve it considerably, we will implement this improvement.
>
I feel the ratings that were present before this sudden sea change were more consistent and valid.
>This way has been honoured in the past, and will continue to be >honoured in the future. This is the only objective way. No individual >feelings, no good-old-days, just facts, facts, facts ...
>
There were never any 'good-old-days', as ratings have been biased for years. However, this current system is insulting to fighters who have actually accomplished a great deal, yet being overlooked in favor of other fighters who simply have not.
What is objective or fact-based about Malik Scott, Tye Fields & Brian Minto being rated ahead of Lyakhovich, Brewster, Byrd, & Williams??? And how can one justify Matt Skelton being ranked ahead of Danny Williams for the time being? If this is a meritocracy, something went horribly wrong.
If we simply want to be mediocre and present ratings that make no logical sense, then we are doing an absolutely superb job.
>Best regards
>
>Martin
Its based on prediction rate over time.. therefore subjectivity is eliminated and objectivity is the only thing left. It is purely statistical now and it will get better.
You won't see any of the shit you see on "official ratings" like Byrd sitting on the top of HW division for years fighting scrubs. None of that. It is based on performance.
>Its based on prediction rate over time.. therefore subjectivity is >eliminated and objectivity is the only thing left. It is purely statistical >now and it will get better.
>
>You won't see any of the shit you see on "official ratings" like Byrd >sitting on the top of HW division for years fighting scrubs. None of that. >It is based on performance.
>
Yeah, and now we'll see Brian Minto scale the ratings instead by fighting scrub minuses, and Chad Dawson assume the top spot by beating an old fringe contender, and Brock do just same by beating the #40 guy.
I'm still curious how Matt Skelton leapfrogged Danny Williams, despite just losing to him and beating a significantly lower ranked tune-up opponent than Williams' to boot. Now that's just mind boggling...
It looks like these ratings simply say which top 20 fighter fought and won last, and that's it, but as the Williams-Skelton example portrays, it's still very, very inconsistent and that reflects very negatively on the validity of such a system.
>
>You won't see any of the shit you see on "official ratings" like Byrd >sitting on the top of HW division for years fighting scrubs. None of that. >It is based on performance.
>
Yeah, and now we'll see Brian Minto scale the ratings instead by fighting scrub minuses, and Chad Dawson assume the top spot by beating an old fringe contender, and Brock do just same by beating the #40 guy.
I'm still curious how Matt Skelton leapfrogged Danny Williams, despite just losing to him and beating a significantly lower ranked tune-up opponent than Williams' to boot. Now that's just mind boggling...
It looks like these ratings simply say which top 20 fighter fought and won last, and that's it, but as the Williams-Skelton example portrays, it's still very, very inconsistent and that reflects very negatively on the validity of such a system.
Diego wrote:>Its based on prediction rate over time.. therefore subjectivity is >eliminated and objectivity is the only thing left. It is purely statistical >now and it will get better.
>
>You won't see any of the shit you see on "official ratings" like Byrd >sitting on the top of HW division for years fighting scrubs. None of that. >It is based on performance.
>
Yeah, and now we'll see Brian Minto scale the ratings instead by fighting scrub minuses, and Chad Dawson assume the top spot by beating an old fringe contender, and Brock do just same by beating the #40 guy.
I'm still curious how Matt Skelton leapfrogged Danny Williams, despite just losing to him and beating a significantly lower ranked tune-up opponent than Williams' to boot. Now that's just mind boggling...
It looks like these ratings simply say which top 20 fighter fought and won last, and that's it, but as the Williams-Skelton example portrays, it's still very, very inconsistent and that reflects very negatively on the validity of such a system.
So.. just because Williams beat Skelton by a narrow split decision.. Williams must be ahead of Skelton for all-time?? How do you handle that situation?
I want specifics too.
U guys are bucking tradition which has always used ratings to determine objectively who deserves through his accomplishments, a shot at the title.
Exactly what purpose BoxRec’s prediction based system serve other than giving publicist a reference for their bogus propaganda?
Shannon Briggs publicist can now point to Boxrec and claim his fighter is 6th rated even though he hasn’t beat a rated boxer in 9 years.
Except for a very few, most of BoxRec’s members and editors are against it. 
U guys need to put on your BoxRec t-shirts and go to shows & see what king of feedback U get. :(
I have stopped wearing mine.
Exactly what purpose BoxRec’s prediction based system serve other than giving publicist a reference for their bogus propaganda?
Shannon Briggs publicist can now point to Boxrec and claim his fighter is 6th rated even though he hasn’t beat a rated boxer in 9 years.
Except for a very few,
most of BoxRec’s members and editors are against it. U guys need to put on your BoxRec t-shirts and go to shows & see what king of feedback U get. :( I have stopped wearing mine.
>So.. just because Williams beat Skelton by a narrow split decision.. >Williams must be ahead of Skelton for all-time?? How do you handle >that situation?
>
>I want specifics too.
>
Uh...yeah, exactly. First of all, what is this about all-time? This should not be about all-time ratings, it is about current rankings. All-time should be in a separate category.
Here's my point, find a hole in it, if you can.
Danny Williams and Matt Skelton fought.
Williams defeated Skelton by close decision, but he won legitimately. Instantly, Williams should be rated above Skelton, however Skelton is rated. In no situation should Skelton be ranked higher than the fighter who just legitimately beat him, no situation at all. In additon, each fighter fought one tune-up fight afterwards. The fighter that Williams beat was ranked higher than Skelton's tune-up opponent, Williams dispatced of him in an earlier round, and Williams fought after Skelton did. Having Skelton ranked higher than Williams prior to the rematch is completely illogical, and I believe even Skelton would even agree to that.
Skelton just beat Williams, and now Williams should legimitimately be behind him in the ratings. This case in point is cut & dry, and it is absolutely ridiculous.
>
>I want specifics too.
>
Uh...yeah, exactly. First of all, what is this about all-time? This should not be about all-time ratings, it is about current rankings. All-time should be in a separate category.
Here's my point, find a hole in it, if you can.
Danny Williams and Matt Skelton fought.
Williams defeated Skelton by close decision, but he won legitimately. Instantly, Williams should be rated above Skelton, however Skelton is rated. In no situation should Skelton be ranked higher than the fighter who just legitimately beat him, no situation at all. In additon, each fighter fought one tune-up fight afterwards. The fighter that Williams beat was ranked higher than Skelton's tune-up opponent, Williams dispatced of him in an earlier round, and Williams fought after Skelton did. Having Skelton ranked higher than Williams prior to the rematch is completely illogical, and I believe even Skelton would even agree to that.
Skelton just beat Williams, and now Williams should legimitimately be behind him in the ratings. This case in point is cut & dry, and it is absolutely ridiculous.