Ratings - please read before commenting - Archived

[computerrank]

hi Martin! I have a question about accrued points in battle
Eleider Alvarez vs Jean Pascal 2017-06-03
referee cards referee:
Woodburn 112-116 | 114-114 Richard DeCarufel | Rodolfo Ramirez 111-117
points before the fight 231-266 after the battle 300-215.
judges notes equal because I think Eleider Alvarez got a lot of points for this fight???

Hi Michael,
Alvarez was in favor with 10 judge score points = 3.33 points per judge on 12 rounds. cd = 3.33/12*2 = 0.56.

So Alvarez got 51 regular points and 18 extra points, because Pascal was #9 before the bout.

[mike1989]

hi Martin! I have a question about accrued points in battle
Eleider Alvarez vs Jean Pascal 2017-06-03
referee cards referee:
Woodburn 112-116 | 114-114 Richard DeCarufel | Rodolfo Ramirez 111-117
points before the fight 231-266 after the battle 300-215.
judges notes equal because I think Eleider Alvarez got a lot of points for this fight???

Hi Michael,
Alvarez was in favor with 10 judge score points = 3.33 points per judge on 12 rounds. cd = 3.33/12*2 = 0.56.

So Alvarez got 51 regular points and 18 extra points, because Pascal was #9 before the bout.

Martin if you follow the logic points should display the official verdict of the judges. According to judges ' notes, fight equal with a small advantage for Alvarez because the difference in points between Pascal and Alvarez can't be much???

[computerrank]

Martin if you follow the logic points should display the official verdict of the judges. According to judges ' notes, fight equal with a small advantage for Alvarez because the difference in points between Pascal and Alvarez can't be much???

The judges score points difference was NOT small.

The difference war 4 + 0 + 6 score points = 3.33 points per judge. In a 12 rounds bout 6 points per judge are regarded a clear decision. 3.33 points are 55 percent of a clear decision.

[mike1989]

Martin if you follow the logic points should display the official verdict of the judges. According to judges ' notes, fight equal with a small advantage for Alvarez because the difference in points between Pascal and Alvarez can't be much???

The judges score points difference was NOT small.

The difference war 4 + 0 + 6 score points = 3.33 points per judge. In a 12 rounds bout 6 points per judge are regarded a clear decision. 3.33 points are 55 percent of a clear decision.

Martin and if it was 36 instead of 10 how many point would get Alvarez ???

[computerrank]

Martin and if it was 36 instead of 10 how many point would get Alvarez ???

Lokk at rule 11 in http://boxrec.com/media/index.php/BoxRe ... escription

"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."

In a bout with 12 rounds boxed the clear decision factor is the score difference per judge / 50% of 12 rounds = score difference per judge / 6

For 36 score points difference - cd = 36 / 3 / 6 = 2.

But for the ratings a maximum of cd of 1 is regarded.

[mike1989]

Martin and if it was 36 instead of 10 how many point would get Alvarez ???

Lokk at rule 11 in http://boxrec.com/media/index.php/BoxRe ... escription

"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."

In a bout with 12 rounds boxed the clear decision factor is the score difference per judge / 50% of 12 rounds = score difference per judge / 6

For 36 score points difference - cd = 36 / 3 / 6 = 2.

But for the ratings a maximum of cd of 1 is regarded.

Martin then is a paradox in MD 3.33 a at maximum UD only 2 ??? and this is 10 rounds vs 36 rounds ???

[computerrank]

Martin then is a paradox in MD 3.33 a at maximum UD only 2 ??? and this is 10 rounds vs 36 rounds ???

In this bout we had a sum of 10 score points difference for all 3 judges This is an average of 3.33 points per judge. And a clear decision factor of cd=0.55.

For a clear decision with 1.0 we have an average difference of 6 points per judge and a difference sum of 18 score points for all 3 judges. More than 18 points difference in sum and more than 6 points difference per judge are not regarded for the ratings.

[StarLord25]

I am lost on what opponents connected means for the additional points. The thread on it in thus topic didn't help me get it and the calculators for the example aren't there anymore

[computerrank]

I am lost on what opponents connected means for the additional points. The thread on it in thus topic didn't help me get it and the calculators for the example aren't there anymore

Boxrec ratings rules 17 and 18 say:
(17) The bout's additional points value comprehends 3 parts
(18) 1 point anyway - another 4 points for an opponent connected by strong results to other connected boxers - and another up to 258 points for an opponent in the division top 50 for men and the division top 30 for women

The formula explanation section says:
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 + 258/(rank-1)) * v * cd * (r_b+10)/(r_a+r_b+20)

A boxer is connected (con=1), if he has wins against or close losses against a connected opponent.

The p4p #1 boxer is the master boxer. He is regarded connected by default.

And then starting with p4p #1
- all boxers connected by wins or close losses against him are regarded connected
- all boxers connected by wins or close losses against these connected boxers are regarded connected
etc until no further boxer can be added

example:
- Francisco Villanueva has a close loss against Alvarez
- Jose de Jesus Macias has a win against Francisco Villanueva
- Ivan Ramirez has a win against Jose de Jesus Macias
So Ivan Ramirez is connected.

[StarLord25]

I am lost on what opponents connected means for the additional points. The thread on it in thus topic didn't help me get it and the calculators for the example aren't there anymore

- all boxers connected by wins or close losses against these connected boxers are regarded connected
etc until no further boxer can be added

Until no further boxer can be added?

[computerrank]

I am lost on what opponents connected means for the additional points. The thread on it in thus topic didn't help me get it and the calculators for the example aren't there anymore

- all boxers connected by wins or close losses against these connected boxers are regarded connected
etc until no further boxer can be added

Until no further boxer can be added?

Boxrec includes about 600,000 boxers currently. About 200,000 of them are connected to the current p4p #1 by a queue of wins or close losses. About 400,000 are not connected to any of these 200,000 connected boxers.

[StarLord25]

So connected has to do with beating an opponent who's fought a top opponent or someone who fought a top opponent?

[computerrank]

So connected has to do with beating an opponent who's fought a top opponent or someone who fought a top opponent?

Yes

[StarLord25]

Thank you.

Another thing is I don't get your calculations. Take DeGale vs Truax at 168 recently. I used your calculations im Excel. On Boxrec, Truax went from 53-160 and DeGale 258-150. I got it calculated as DeGale: 258-187 and Truax 53-149. What went wrong with my calculation or what's different.?

[computerrank]

Thank you.

Another thing is I don't get your calculations. Take DeGale vs Truax at 168 recently. I used your calculations im Excel. On Boxrec, Truax went from 53-160 and DeGale 258-150. I got it calculated as DeGale: 258-187 and Truax 53-149. What went wrong with my calculation or what's different.?

What was your calculation?

[m1kee50]

Martin, when was the last 'big 4' workd title fight at York Hall please?

[computerrank]

Martin, when was the last 'big 4' workd title fight at York Hall please?

http://boxrec.com/en/event/130337

Code: Select all

show_id 	title 	date
130337 	WBO 	2002-01-19
44410 	WBO 	2001-01-27
17315 	WBO 	2000-04-08
40346 	WBO 	1995-10-21
31666 	WBO 	1994-07-30
5836 	WBO 	1994-06-11

[m1kee50]

Thanks Martin, much obliged.

[StarLord25]

Thank you.

Another thing is I don't get your calculations. Take DeGale vs Truax at 168 recently. I used your calculations im Excel. On Boxrec, Truax went from 53-160 and DeGale 258-150. I got it calculated as DeGale: 258-187 and Truax 53-149. What went wrong with my calculation or what's different.?

What was your calculation?

I use your formula and got the values I quoted about instead of the ones that are there on the site.

[computerrank]

Thank you.

Another thing is I don't get your calculations. Take DeGale vs Truax at 168 recently. I used your calculations im Excel. On Boxrec, Truax went from 53-160 and DeGale 258-150. I got it calculated as DeGale: 258-187 and Truax 53-149. What went wrong with my calculation or what's different.?

What was your calculation?

I use your formula and got the values I quoted about instead of the ones that are there on the site.

- what formula did you use?
- what values did you use with it?

[StarLord25]

The same one on the ratings description page and fighter A being Truax and B being Degale. I use the values they had prior to their fight.

[computerrank]

The same one on the ratings description page and fighter A being Truax and B being Degale. I use the values they had prior to their fight.

The winner is always rated higher than the loser. This rule wasn't documented in the description. I added it as rule 26.

[sinisap]

Hello,
I appreciate your complex ratings points calculation attempts
But:
Chris Eubank Jr = 423
Billy Joe Saunders = 230
so it looks like that formulas are in need of review and redesign ....
All the best

[computerrank]

Hello,
I appreciate your complex ratings points calculation attempts
But:
Chris Eubank Jr = 423
Billy Joe Saunders = 230
so it looks like that formulas are in need of review and redesign ....
All the best

Saunders defeated Eubank in a very close decision 2014.

But the quality of the following results make the difference:
- Eubank then stopped Chudnikov with 140 points, Blackwell with 105 points, Quinlan with 157 points, Yildrim with 171 points and clearly defeated Abraham with 254 points.
- Saunders then only had a close MD to Lee with 242 points and a close win against Monroe with 155 points.

[DaveyMac]

Hello,
I appreciate your complex ratings points calculation attempts
But:
Chris Eubank Jr = 423
Billy Joe Saunders = 230
so it looks like that formulas are in need of review and redesign ....
All the best

There is no rating system in the world where one can't cherry pick some perceived inconsistency. I don't think the system should be "redesigned" every time you find one.

I've done exhaustive research on the Boxrec rating system, and although I think it was slightly more accurate a couple of jumps back, it is by far the best predictor of success one can use. Damn fine system if you ask me. :)