Riddle me this

[oogiebe]

So BoxRec changes the HW rankings AFTER today's Bellow (made me look bad) ko of Haye. It snuck in as I had viewed earlier today.

Haye gets inserted into #15
Bellew gets inserted into #4.

Nonsense for both.

[jamamb]

they havent even given bellew his points from todays win either, he could end up even higher then 4

[oogiebe]

they havent even given bellew his points from todays win either, he could end up even higher then 4

Ok, so it's not just me then.

[BitPlayer]

The rankings are always a bit funny, but they've went especially bad as of late.

James Toney is a top 10 all time heavyweight
Sam Langford is 183rd all time P4P
Harry Greb is not even in the top 10 Middleweight rankings
Max Baer is above Lennox Lewis and Mike Tyson
Tomasz Adamek and Marvin Hart are above Jack Dempsey
Glazgov and Valuev are above Jersey Joe Wallcot
Ted Kid Lewis is behind some random guy with only one fight

[Grant]

Think Max Baer got there on quantity. Hell of a career though when you look at it

[gilgamesh]

So BoxRec changes the HW rankings AFTER today's Bellow (made me look bad) ko of Haye. It snuck in as I had viewed earlier today.

Haye gets inserted into #15
Bellew gets inserted into #4.

Nonsense for both.

I don't know how the rankings are compiled, but frankly Boxrec's rankings have always been a joke.

It's a record data base. It's not even one of the 10th most reliable authorities on rankings.

[oogiebe]

Think Max Baer got there on quantity. Hell of a career though when you look at it

Read the thread from the start, please. Wasn't the point, although your point is well taken.

[oogiebe]

they havent even given bellew his points from todays win either, he could end up even higher then 4

Jamb - Look at the last two Bellew fight points accumulations.

[oogiebe]

So BoxRec changes the HW rankings AFTER today's Bellow (made me look bad) ko of Haye. It snuck in as I had viewed earlier today.

Haye gets inserted into #15
Bellew gets inserted into #4.

Nonsense for both.

I don't know how the rankings are compiled, but frankly Boxrec's rankings have always been a joke.

It's a record data base. It's not even one of the 10th most reliable authorities on rankings.

This time it's really nonsensical.

[gilgamesh]

So BoxRec changes the HW rankings AFTER today's Bellow (made me look bad) ko of Haye. It snuck in as I had viewed earlier today.

Haye gets inserted into #15
Bellew gets inserted into #4.

Nonsense for both.

I don't know how the rankings are compiled, but frankly Boxrec's rankings have always been a joke.

It's a record data base. It's not even one of the 10th most reliable authorities on rankings.

This time it's really nonsensical.

Name me a time when it wasn't nonsensical

[oogiebe]

I don't know how the rankings are compiled, but frankly Boxrec's rankings have always been a joke.

It's a record data base. It's not even one of the 10th most reliable authorities on rankings.

This time it's really nonsensical.

Name me a time when it wasn't nonsensical

...ok...good point...!

[Wales]

So BoxRec changes the HW rankings AFTER today's Bellow (made me look bad) ko of Haye. It snuck in as I had viewed earlier today.

Haye gets inserted into #15
Bellew gets inserted into #4.

Nonsense for both.

I don't know how the rankings are compiled, but frankly Boxrec's rankings have always been a joke.

It's a record data base. It's not even one of the 10th most reliable authorities on rankings.

all these ratings evaluate every day all bouts in the database in chronological sequence. A higher rated boxer should be expected to defeat a lower rated boxer with increasing probability by increasing rating difference.

Current Ratings
Every boxer gets a first rating of 0 before his first bout.
After every bout, the ratings of the two boxers involved are changed depending on the bout's official result (KO, TKO, RTD, UD, PTS, NWS, MD, SD, DQ, TD, DRAW).
The value of a result varies between v=1 and v=0.
The clear decision factor varies between cd=1 and cd=0.
The winner cannot lose points for KO, TKO, RTD, DQ, TD and decisions on points with cd=1
KO, TKO, RTD are rewarded with full value v=1, cd=1.
NWS is rewarded with full value v=1 for 12 rounds boxed and more and a lower value related to the number of rounds boxed. Clear decision factor cd=1.
UD, PTS, DQ, TD are rewarded with full value v=1 for 12 rounds boxed and more and a lower value related to the number of rounds boxed, clear decision factor cd=1. If the score cards are available, it may be less.
MD, SD are rewarded with full value v=1 for 12 rounds boxed and more and a lower value related to the number of rounds boxed. Clear decision factor is limited to cd=0.5. If the score cards are available cd may be less.
DRAW is rewarded with full value v=1 for 12 rounds boxed and more and a lower value related to the number of rounds boxed. Clear decision factor cd=0.
If the score cards are available, the value rewarded is in direct proportion to the rounds boxed, with full value v=1 for 12 rounds boxed and more. The clear decision factor is in proportion to rounds boxed and the mean score difference per judge. cd=1 for a mean score difference per judge of 50% of the rounds boxed.
All bouts are regarded to have the same weight independent of titles.
The winner gets a certain part of the opponent's points and a certain part of the rating difference to the opponent's rating.
For a DRAW the rating of the higher rated boxer is reduced by some part of the point difference; the rating of the lower rated boxer is enhanced by the same amount of points.
The full relative point reward is 33%. It is in direct proportion to the pre-bout rating of the defeated opponent.
The winner also gets additional points from the bout's additional points value in proportion of the opponent's rating to his own rating. The proportion is (r_b+10) / (r_a+r_b+20), where r_a is the pre-bout rating of the winner and r_b is the pre-bout rating of the loser.
The bout's additional points value comprehends 3 parts
1 point anyway - another 4 points for an opponent connected by strong results to other connected boxers - and another up to 250 points multiplied by bout value v and clear decision factor cd for an opponent in the division top 50 for men and the division top 30 for women
The rating of a boxer is reduced, if he didn't box an opponent with a rating of at least 50% (for men) or 33% (for women) of his own rating points within 18 months.
The rating of a boxer is reduced by up to 50% in proportion to the difference of 2 times the rating points of his best opponent in this time period minus his own rating for men.
The rating of a boxer is reduced by up to 50% in proportion to the difference of 3 times the rating points of his best opponent in this time period minus his own rating for women.
The reduction is in proportion to the time the requirement was missed.
The rating of a boxer is reduced by 50% for every time period of inactivity of 18 months.
The pre-bout rating of a successfully debuting boxer is set to 25% of his opponents pre-bout rating.
The rating points are in relation to a weight division. The rating points are converted with the cube of the upper weight limit ratio of the old and new weight division.
The winner is always rated higher than the loser. the minimum margin is: mean pre-bout ratings of both r_m = (r_a + r_b)/2; r_new_winner = r_m + earn_f*v*cd/4; r_new_loser= r_m - earn_f*v*cd/4
Formula

If a boxer with a rating of r_a before the fight defeats a boxer b with a rating of r_b before the fight with result of value v and clear decision factor cd, the new ratings r_a_new and r_b_new after a fight are, earn_f is 33.3%:

earn = earn_f * v * (r_b*cd + (r_b-r_a)/(1+2*cd));
r_a_new = r_a + earn
r_b_new = r_b - earn
Additional points:

opponent is connected or not: con=(1 or 0), value of result v, clearness of decision cd, winner's rating r_a, opponents rating r_b, opponent's rank in division rank (#1 is equivalent to #2):
additional points winner = (1 + 4*con + 250/(rank-1) * v * cd) * (r_b+10)/(r_a+r_b+20)


Rating reduction caused by missing opponent quality:

r_red-10 = (r_old-10) * (1 - 0.5*(1 - 2*best_opp/r_old)) for men
r_red-10 = (r_old-10) * (1 - 0.5*(1 - 3*best_opp/r_old)) for women

Examples


Boxer a UD 6 boxer b, scores 59:55 58:56 58:56, a has 1000 points, b has 500 points, opponent's rank outside top 50 and he is not connected

A 6 rounder is rewarded with value 6/12, v=0.5
UD is rewarded with cd=1 at maximum
mean score difference per judge is (4+2+2)/3 = 2.667, which is rewarded in direct proportion to half the rounds boxed with cd= 2.667/3 = 0.89 at maximum
so cd=0.89
earn= 0.33 * 0.5 * (500*0.89 + (500-1000)/(1+2*0.89)) = 44
r_a_new = 1000 + 44 = 1044
r_b_new = 500 - 44 = 456
additional points:
additional points winner = (1 +4*0)* (500+10)/(500+1000+20) = 0
r_a_new = 1044 + 0 = 1044
r_b_new = 456 + 0 = 456

Boxer a SD 4 boxer b, scores 39:37 39:37 37:39, a has 1000 points, b has 500 points, opponent's rank outside top 50, but he is connected

A 4 rounder is rewarded with 4/12, v=0.333
SD is rewarded with cd=0.5 at maximum
mean score difference per judge is (2+2-2)/3 = 0.667, which is rewarded in direct proportion to half the rounds boxed 0.667/2 at maximum\
so cd=0.333
earn= 0.33 * 0.33 * (500*0.33 + (500-1000)/(1+2*0.33)) = -15
r_a_new = 1000 - 15 = 985
r_b_new = 500 + 15 = 515
additional points:
additional points winner = (1 +4*1)* (500+10)/(500+1000+20) = 2
r_a_new = 985 + 2 = 987
r_b_new = 515 + 0 = 515

More complex - Boxer a KO 4 boxer b, a has 300 points, b has 400 points, boxer b is connected con=1, v=1, cd=1, opponent's rank in division rank=5

earn= 0.33 * 1 * (400*1 + (400-300)/(1+2*1)) = 143
r_a_new = 300 + 143 = 443
r_b_new = 400 - 143 = 257
additional points:
additional points winner = (1 +4*1 + 250/(5-1) * 1 * 1)* (400+10)/(400+300+20) = 38
r_a_new = 443 + 38 = 481
r_b_new = 257 + 0 = 257

[oogiebe]

Ohhhh...thanks for clearing that up for me! lmao!

[Wales]

Good luck making sense of that. In short it’s a mathematical computerised formula that adjusts the ratings, there’s no human element.

[oogiebe]

Good luck making sense of that. In short it’s a mathematical computerised formula that adjusts the ratings, there’s no human element.

Yeah...you made your point. LMAO! Great response! I'm still laughing! It's great having the math, but they should put some eyes on this. You response was PRICELESS!

[rab]

in other words it is complete nonsense. maths doesn't work when the object of the exercise is subjective.

[oogiebe]

in other words it is complete nonsense. maths doesn't work when the object of the exercise is subjective.

Wha...?

[Wales]

Guess there’s no way it could be done any other way as one person couldn’t watch every fight , hundreds are not even recorded let alone televised.

[oogiebe]

Guess there’s no way it could be done any other way as one person couldn’t watch every fight , hundreds are not even recorded let alone televised.

Agreed, however; how much effort scanning over the top 100 or even 50 to ensure that Pareto's Pyramid is somewhat reasonable (I.E. 20% of the list is 80% of the viewed data. It is simply one more step in the process and tell me how long would that take? (tic toc tic toc...) Ok I just did it! Just a review of the top 50 in each division including ATG list etc. Or create an exception report that shows any move over 3 spots up or down after after calculating and before uploading.

How much is it worth to have integrity in an area that sorely needs it...Boxing Divisional Ratings!!! IT JUST DOESN'T MATTER, IT JUST DOESN'T...OH...sorry.

Whew!!!! Ok...I'm better now.

[candyslim]

The ratings are computed using a formula. Apart from the formula itself there is no human influence on the resulting order.

The advantage is that the ratings are free from subjective bias. The disadvantage is that the formula - any formula - is imperfect and throws up some very strange anomolies.

I like the boxrec ratings especially how comprehensive they are. I know not to take them as gospel and I'm very aware of the flaws, but I see none that are better. If they appeal to me i.e. they are similar to my own ideas at the time, then the chances are they are only a top ten or fifteen.

[Lackeos]

You're never going to get ratings that are totally accurate and totally objective and go down into the top few hundred ranks.

If you want halfway accurate, you can look at The Ring and ESPN, but they aren't even that accurate, and they aren't objective. In fact, "the human element" that keeps getting mentioned in this thread is literally subjectivity.

I've seen countless humans post their rankings on here who happened to be totally out of their mind and included anomalies such as Chad Dawson #1 p4p or Vasyl Lomachenko top 4 all-time. In other words, you really can't trust a human's opinion either, unless they happen to be in the top 1 percentile of best rankers. And I hate to think of how inaccurately a human would turn-in a top 150 rankings list if they didn't at least have a computerized rankings to refer to as a guideline.

I like having an objective and deep rankings list to refer to. I'd hate to be debating the ballpark #100 opponents that are being fought by a roughly #40-rated fighter and just have to use "our own personal rankings" when arguing with someone non-compliant. I could say this opponent is #50 in my rankings, and the other person could say the opponent is #1050 in their rankings. Considering that I recently read someone on boxrec labeling Herbe Herbeaux as a "bum," and given that bum status equates to about a #700 ranking, posters on boxrec aren't shy about labeling a boxer several hundred ranks below what he really deserves. So you think an opponent is #50, the other guy thinks the opponent is #1050, and what objective source do you have to refer to (that goes that deep in the ranks) to sort out who is right?

In fact, I'd say that Bellew - Haye is not a very intellectually honest example of boxrec's inaccuracy. David Haye is a fighter who could be #3 when healthy (which he never will be healthy and prime again). Bellew and Haye fought twice in a row, and Haye was injured both times. There is no way you could possibly deceive the formula more than to have Haye injured both times, lose both times, and not tell the formula that he's injured. That's totally twisted. That's like blindfolding someone, telling them to hit a pinata, and then tripping them and making fun of them for poor balance. I mean, you blindfolded the formula's ability to know whether Haye was injured, you tripped it up by Haye being injured, and then you mock it for not accurately ascertaining what's what in that fight.

If we could tag fights as having an injured fighter, and if judges submitted accurate scorecards (or if the community could override the official scorecards in some high profile cases), I believe the rankings accuracy would be remarkably improved. For instance, Adamek would have had a lot lower ranking from beating an injured Chambers and robbing Cunningham in the rematch. Then he wouldn't have given as big of a boost to Szpilka, Molina and Glazkov. Breazeale wouldn't have gained as many points from the inflated Molina ranking, nor as many points from the tongue-bitten Mansour. Kownacki wouldn't have gained as many points from Szpilka. Charles Martin wouldn't have gained as many points from Glazkov, both because Glazkov's rating would be lower and because the formula would know about Glazkov's injured. And that's just a small sample of the repercussions that would originate from just Tomasz Adamek's opponents.

[squiggy]

I think that, love it or hate it, it's a mathematical system, and anything you do to subjectively tweak it because you don't like the results is in serious danger of changing it into just a system that uses a formula on its way to lining up the fighters in the way that seems right to you. In which case, just bypass all the numbers and write the list that you like.

[oogiebe]

I think that, love it or hate it, it's a mathematical system, and anything you do to subjectively tweak it because you don't like the results is in serious danger of changing it into just a system that uses a formula on its way to lining up the fighters in the way that seems right to you. In which case, just bypass all the numbers and write the list that you like.

Agreed. A pair of eyes on the top 50 would help keep some integrity rather than 100% dependency on the equations.

[candyslim]

I think what squiggy is saying Oogie, is that you shouldn't tamper with it to correct these anomolies that Lackeos is talking about, because when you do that you are introducing human subjectivity into it.

I can see where he's coming from because although on the face of it, you are righting a wrong, in effect you are taking it away from being a dispassionate and 'honest' mathematical exercise, and turning it into something generated or at least influenced by opinion, and then it becomes a slippery slope so to speak.

Good post though Lackeos.

[Ilya Muromets]

So BoxRec changes the HW rankings AFTER today's Bellow (made me look bad) ko of Haye. It snuck in as I had viewed earlier today.

Haye gets inserted into #15
Bellew gets inserted into #4.

Nonsense for both.

They've got Bellew #4 and Haye 19 now. Still nonsense.