Ratings - please read before commenting - Archived

[mike1989]

Hi Michael,

- Is there a website, where the results are published?
- is there a newspaper report with the results?

I cannot find an official homepage of National Professional Boxing League of Ukraine

Greetings
Maretin
[/quote]

no website League of professional Boxing of Ukraine, the tournament in Kharkov 17.12.2017 was a small club level.Little information I asked them for the results that I was told that registration in boxrec refused.
Greetings
Mikhail

[computerrank]

Hi Michael,

- Is there a website, where the results are published?
- is there a newspaper report with the results?

I cannot find an official homepage of National Professional Boxing League of Ukraine

Greetings
Maretin

no website League of professional Boxing of Ukraine, the tournament in Kharkov 17.12.2017 was a small club level.Little information I asked them for the results that I was told that registration in boxrec refused.
Greetings
Mikhail
[/quote]
Hi Michael,
I sent you a PM
Greetings
Martin

[giacomino]

Question, and I apologize in advance if this had been asked and answered 123 times already: are there extra points given for an opponent or belt-holder winning in their opponent’s home country? Came across that when I noticed Brian Castaño being ranked behind Michel Soro at 154, despite having beaten him in his most recent fight. I understand Soro has fought much stiffer opposition in the past, so thought that might be the reason, but seemed odd since Castaño beat the Frenchman in France and is undefeated.

[jamamb]

boxrec has had stuff like that before where the head 2 head losing boxer is very shortly after ahead

imo boxrec should have something in place specifically to account for head to head results in the later rankings, to prevent situations like that. like maybe for a certain number of fights after the head 2 head, the winning boxer also gets extra points based on increases the losing boxer has. thats just an example.

cant imagine that anyone really thinks soro should already be back ahead of castano. there is certainly a point where the value of a head to head win diminishes , but not this fast

[computerrank]

@giacomo and jamamb

Brian Castaño vs Michel Soro was a very close SD., Danilo Dongo 115-113 | Pasquale Procopio 115-113 | Luigi Boscarelli 112-116
and Castrano had only a third of Soro's points before the bout.

So Castrano was in front by only 1 point - and a small achievement could exchange the ranks.

From my perspective binding a winner and a loser is not a good idea. Every bout and every match-up is different - and achievements should be rated independently.

e.g. if a winner of A vs B loses to C - maybe in an upset - B should'nt suffer from this. And if the loser of A vs B wins against C - A shouldn't sponge points for doing nothing.

[Manrae]

WHAT JUST HAPPENED TO THE RATINGS???

Can someone explain the new changes please??

Zab Judah is now ranked above Henry Armstrong in the ALL-TIME Rankings

[computerrank]

WHAT JUST HAPPENED TO THE RATINGS???

Can someone explain the new changes please??

Zab Judah is now ranked above Henry Armstrong in the ALL-TIME Rankings

- They no longer get any points for just staying in the tops (no points for annual division end ranks)
- They only get annual points for defeating their best ranked opponent - limited by the rank of their worst opponent lost to in this year

Code: Select all

    id|division                      |  year|     r|at_sum|at_ann|at_opp|  at_r|name                          

------|------------------------------|------|------|------|------|------|------|------------------------------

  6945|Light Welterweight            |  1998|   155|    29|     0|    17|    12|Zab Judah                     

  6945|Light Welterweight            |  1999|   153|    30|     0|    17|    13|Zab Judah                     

  6945|Light Welterweight            |  2000|   657|   242|     0|   217|    26|Zab Judah                     

  6945|Light Welterweight            |  2001|   446|   242|     0|   217|    26|Zab Judah                     

  6945|Light Welterweight            |  2002|   452|   256|     0|   230|    26|Zab Judah                     

  6945|Light Welterweight            |  2003|   375|   273|     0|   247|    26|Zab Judah                     

  6945|Welterweight                  |  2004|   277|   279|     0|   254|    26|Zab Judah                     

  6945|Welterweight                  |  2005|   809|   482|     0|   454|    28|Zab Judah                     

  6945|Welterweight                  |  2006|   354|   482|     0|   454|    28|Zab Judah                     

  6945|Light Middleweight            |  2007|   206|   482|     0|   454|    28|Zab Judah                     

  6945|Welterweight                  |  2008|   220|   482|     0|   454|    28|Zab Judah                     

  6945|Welterweight                  |  2009|   220|   482|     0|   454|    28|Zab Judah                     

  6945|Light Welterweight            |  2010|   181|   489|     0|   461|    28|Zab Judah                     

  6945|Light Welterweight            |  2011|   249|   489|     0|   461|    28|Zab Judah                     

  6945|Light Welterweight            |  2012|   277|   502|     0|   474|    28|Zab Judah                     

  6945|Light Welterweight            |  2013|   178|   502|     0|   474|    28|Zab Judah                     

Code: Select all

    id|division                      |  year|     r|at_sum|at_ann|at_opp|  at_r|name                          

------|------------------------------|------|------|------|------|------|------|------------------------------

  9018|Lightweight                   |  1933|   177|    13|     0|     0|    13|Henry Armstrong               

  9018|Featherweight                 |  1934|   566|    24|     0|     0|    24|Henry Armstrong               

  9018|Featherweight                 |  1935|   940|    31|     0|     0|    31|Henry Armstrong               

  9018|Featherweight                 |  1936|   755|    36|     0|     0|    36|Henry Armstrong               

  9018|Lightweight                   |  1937|  1948|   244|     0|   200|    44|Henry Armstrong               

  9018|Welterweight                  |  1938|  2118|   448|     0|   400|    48|Henry Armstrong               

  9018|Lightweight                   |  1939|  2228|   481|     0|   433|    48|Henry Armstrong               

  9018|Heavyweight                   |  1940|   569|   496|     0|   448|    48|Henry Armstrong               

  9018|Welterweight                  |  1941|  1093|   496|     0|   448|    48|Henry Armstrong               

  9018|Welterweight                  |  1942|  1517|   496|     0|   448|    48|Henry Armstrong               

  9018|Welterweight                  |  1943|  1175|   496|     0|   448|    48|Henry Armstrong               

  9018|Welterweight                  |  1944|   696|   496|     0|   448|    48|Henry Armstrong               

  9018|Welterweight                  |  1945|   372|   496|     0|   448|    48|Henry Armstrong               

[JCS]

- They only get annual points for defeating their best ranked opponent - limited by the rank of their worst opponent lost to in this year

This is incredibly biased against the old-school boxers....

[computerrank]

- They only get annual points for defeating their best ranked opponent - limited by the rank of their worst opponent lost to in this year

This is incredibly biased against the old-school boxers....

I don't think so. They defeated each other occasionally, and they lost to each other occasionally too. So some of them got a lot of points, although they performed quite questionable.

[JCS]

- They only get annual points for defeating their best ranked opponent - limited by the rank of their worst opponent lost to in this year

This is incredibly biased against the old-school boxers....

I don't think so. They defeated each other occasionally, and they lost to each other occasionally too. So some of them got a lot of points, although they performed quite questionable.

So... let me see if I understand.

Let's say my year 2017 goes like this.

I beat 4000 rated fighter
I lose to 150 rated fighter
I beat 2500 rated fighter (in a rematch, this is the 150 previous who beat me)
I beat a 5000 rated fighter


Are you saying the 150 loss (which I later avenged) cancels out the 3 wins ?

[computerrank]

This is incredibly biased against the old-school boxers....

I don't think so. They defeated each other occasionally, and they lost to each other occasionally too. So some of them got a lot of points, although they performed quite questionable.

So... let me see if I understand.

Let's say my year 2017 goes like this.

I beat 4000 rated fighter
I lose to 150 rated fighter
I beat 2500 rated fighter (in a rematch, this is the 150 previous who beat me)
I beat a 5000 rated fighter


Are you saying the 150 loss (which I later avenged) cancels out the 3 wins ?

Yes, if it happens the same year.

[JCS]

I don't think so. They defeated each other occasionally, and they lost to each other occasionally too. So some of them got a lot of points, although they performed quite questionable.

So... let me see if I understand.

Let's say my year 2017 goes like this.

I beat 4000 rated fighter
I lose to 150 rated fighter
I beat 2500 rated fighter (in a rematch, this is the 150 previous who beat me)
I beat a 5000 rated fighter


Are you saying the 150 loss (which I later avenged) cancels out the 3 wins ?

Yes, if it happens the same year.

Seems illogical

[computerrank]

So... let me see if I understand.

Let's say my year 2017 goes like this.

I beat 4000 rated fighter
I lose to 150 rated fighter
I beat 2500 rated fighter (in a rematch, this is the 150 previous who beat me)
I beat a 5000 rated fighter


Are you saying the 150 loss (which I later avenged) cancels out the 3 wins ?

Yes, if it happens the same year.

Seems illogical

The logic is to select stable superiority.

[JCS]

Yes, if it happens the same year.

Seems illogical

The logic is to select stable superiority.

I get the idea, but there are more holes in this method than in swiss cheese.

[computerrank]

Seems illogical

The logic is to select stable superiority.

I get the idea, but there are more holes in this method than in swiss cheese.

I like swiss cheese

For the all time ratings the method convinces me most now, given superioritity over many years as the measure.

Where do you see the holes?

[JCS]

The logic is to select stable superiority.

I get the idea, but there are more holes in this method than in swiss cheese.

I like swiss cheese

For the all time ratings the method convinces me most now, given superioritity over many years as the measure.

Where do you see the holes?

I think the holes will be apparent once analyzing some year-to-year totals for boxers from the past..

What does Sugar Ray Robinson's 1955 look like?

Admittedly, it is harder to find examples like this than I thought.

How would the results change if you allowed the one worst loss in a year per fighter to not count in this limitation?

Win against 4500
Win against 2000
Loss against 350
Loss against 1200

Limitation is 1200 this year. (350 is exempt)


Or how about the worst the limitation can be, is one-half of the best win.

Win against 4500
Win against 2000
Loss against 350
Loss against 1200

Worst loss is 350.. but that exceeds 1/2 of 4500... so therefore, the limitation is 2250. Or maybe try one-third.... or one-quarter.


Henry Armstrong being above Zab Judah should shout... something is wrong.

[Manrae]

The logic is to select stable superiority.

I get the idea, but there are more holes in this method than in swiss cheese.

I like swiss cheese

For the all time ratings the method convinces me most now, given superioritity over many years as the measure.

Where do you see the holes?

Mathematical algorithms aside...

Can you argue Zab Judah being greater than Henry Armstrong all-time? Whether career accomplishments, in-ring skills, historical impact, or a mixture of factors...

[JCS]

I get the idea, but there are more holes in this method than in swiss cheese.

I like swiss cheese

For the all time ratings the method convinces me most now, given superioritity over many years as the measure.

Where do you see the holes?

Mathematical algorithms aside...

Can you argue Zab Judah being greater than Henry Armstrong all-time? Whether career accomplishments, in-ring skills, historical impact, or a mixture of factors...

Gotta agree here. I think Martin will be reasonable and make adjustments so old school boxers are not so heavily penalized in comparison to current day.

I don't have the exact statistic, but top fighters probably have about an 85% win percentage (or so) once they hit their prime. If said fighter fights 15 times per year, like an Armstrong, statistics suggest he'll probably lose 1 of them. In 2018, we are lucky if top fighters fight 3 times per year. Chances are, they'll go a few years without losing, which is something that we see all the time.

If I'm reading his table correctly, Armstrong stopped gaining "all-time" points in 1940, which seems ridiculous. In 1942, for example, he went 13-1. He lost to Reuben Shank, who had 313 points... but beat several fighters above 313, most notably, Fritzie Zivic who was at 1,038.

There could definitely be some smarts in this... You could say something such as... for every 5 wins per calendar year, 1 loss is exempted... starting with the worst loss, 2nd worst... etc. This way, a modern era fighter will be punished for any loss.. while a guy like Armstrong could be excused here and there.

[computerrank]

I like swiss cheese

For the all time ratings the method convinces me most now, given superioritity over many years as the measure.

Where do you see the holes?

Mathematical algorithms aside...

Can you argue Zab Judah being greater than Henry Armstrong all-time? Whether career accomplishments, in-ring skills, historical impact, or a mixture of factors...

....
There could definitely be some smarts in this... You could say something such as... for every 5 wins per calendar year, 1 loss is exempted... starting with the worst loss, 2nd worst... etc. This way, a modern era fighter will be punished for any loss.. while a guy like Armstrong could be excused here and there.

Maybe there is a way to weaken the rigid limit with the worst loss per year. I will go into it.

[JCS]

Mathematical algorithms aside...

Can you argue Zab Judah being greater than Henry Armstrong all-time? Whether career accomplishments, in-ring skills, historical impact, or a mixture of factors...

....
There could definitely be some smarts in this... You could say something such as... for every 5 wins per calendar year, 1 loss is exempted... starting with the worst loss, 2nd worst... etc. This way, a modern era fighter will be punished for any loss.. while a guy like Armstrong could be excused here and there.

Maybe there is a way to weaken the rigid limit with the worst loss per year. I will go into it.

Cool.. interested in seeing potential results. Sam Langford may be another guy to take a look at. He seems pretty low now as well.

[mike1989]

why did Vasilia Lomachenko have a rating of 925 points and became 814 points ???

[computerrank]

why did Vasilia Lomachenko have a rating of 925 points and became 814 points ???

His last bout matching 50% of his rating was against Nicholas Walters on 2016-11-27,

18 months later on 2018-05-27 his best opponent Jason Sosa only had 359 points. This was less than 50% of Lomchenko's 925.

[jamamb]

wow, he moves up and beats linares and gets a poor opposition penalty? i guess thats the problem when you have so many points, even the best guy in the division you can fight (who is rated at that weight) still doesnt cut it then

[JCS]

wow, he moves up and beats linares and gets a poor opposition penalty? i guess thats the problem when you have so many points, even the best guy in the division you can fight (who is rated at that weight) still doesnt cut it then

This opposition penalty has always felt harsh to me.... I think a 1/3rd cutoff is probably more appropriate than the 1/2.

Martin -- Any update on the all-time situation that was brought up?

[computerrank]

wow, he moves up and beats linares and gets a poor opposition penalty? i guess thats the problem when you have so many points, even the best guy in the division you can fight (who is rated at that weight) still doesnt cut it then

This opposition penalty has always felt harsh to me.... I think a 1/3rd cutoff is probably more appropriate than the 1/2.

Martin -- Any update on the all-time situation that was brought up?

- A 1/3 cut off is in place for the women - but there is much less competition.
Boxers at Lomachenkov's level have to meet the absolute tops - there are Mikey Garcia and Miguel Berchelt left.
- Will need some time for investigating the all-time situation.