BoxRec boxers ratings recalculation?

[imaioral]

Can someone tell me why new recalculation? what are the new parameters and/or variables now? And another thing, why Canelo is always with the same rating as before? in last 2 calculations "upgrades" while others decreased their rating he was increased, why? maybe some boxrec fan protecting him?...

[Ruthless-RKO]

maybe weight class made a difference?

[imaioral]

maybe weight class made a difference?

And didn't other boxers went up weight classes too?

[Ruthless-RKO]

maybe weight class made a difference?

And didn't other boxers went up weight classes too?

Im not saying thats the reason, but i do see lets say a fighter at 147 has a rating of 250, his next fight is at 154 so his rating reduces to 230, he then loses that fight, lets say his rating goes down up to 200, but then his next fight is back at 147, so his rating goes back up to 250.

I don't know if that links to what you're referring to, but I admit, i do see what you're saying. I'm just like you trying to figure it out

[imaioral]

It's sad, the minimum they should do is to post an official announcement telling the news about what was done with new ratings calculations for us don't stay like dumbs trying to figure out what is going on about what they did...

[greg]

http://boxrec.com/forum/viewtopic.php?f ... start=2875

[imaioral]

http://boxrec.com/forum/viewtopic.php?f ... start=2875

I read that before, but seems not too much in detail explanatory.

[computerrank]

http://boxrec.com/forum/viewtopic.php?f ... start=2875

I read that before, but seems not too much in detail explanatory.

The ratings description will be updated soon ...

[Enlightened-One]

Can someone tell me why new recalculation? what are the new parameters and/or variables now? And another thing, why Canelo is always with the same rating as before? in last 2 calculations "upgrades" while others decreased their rating he was increased, why? maybe some boxrec fan protecting him?...

Follow this thread:

"New Ratings Release 29 Live"

A new rating release has been launched.

The new ratings considerably improve the ranking precision measured by the corrrect winner prediction of the bouts.

The error ratio of the old ratings bout winner predictions was 26 percent higher than the error ratio of the new ratings. Old 22.5 percent errors - new 17.9 percent errors.

This is another contribution to Boxrec ratings objectivity.

- new additional point procedure depending on
-- opponent's connectedness to valid boxers
-- pre-bout rank of opponent
-- pre-bout rating ratio of opponent to boxer

- bug fixes within missing opponent quality and inactivity points reduction

[imaioral]

http://boxrec.com/forum/viewtopic.php?f ... start=2875

I read that before, but seems not too much in detail explanatory.

The ratings description will be updated soon ...

Thanks 'computerrank'

[Lennox]

Massive change in the ratings. So yesterday Boxrec ratings were shit and today they are good?

No Yesterday they were shit and today they are shittier. On what possible objective grounds could Anthony Joshua be number 1?

[computerrank]

Massive change in the ratings. So yesterday Boxrec ratings were poo and today they are good?

No Yesterday they were poo and today they are shittier. On what possible objective grounds could Anthony Joshua be number 1?

Because heavyweight is one of the least competitive divisions currently. No heavyweight in the top 20 p4p now - Joshua is #40 ...

the last 2 years
- Joshua defeated Martin at 244 points
- Wilder defeated Stiverne at 181 points
- Ortiz defeated Scott at 135 points
- Pulev defeated Chisora at 101 points
- Povetkin defeated Duhaupas at 101 points

Lennox, you are the greatest

[HomicideHenry]

Just proof positive that BoxRec's rankings are democractically driven..... since all the hype is on Joshua and there are so many willing contributors to the site already giving the Englishman a rim job before he's ever proven himself on any real scale..... and because so many posters who vote are also on the Joshua bandwagon.... the ratings cannot be ever taken seriously.... It'd be like if an influx of Marciano fans became members on the site and voted him as the #1 heavyweight of all time, and knocking Ali down to number two.

[Azmir]

Charles Martin over Parker and Haye? Really? Other thing I can't understand is why' Glazkov is treated so well, man he's over Martin he lost to. Even Lucas Browne deserves to be ranked over Martin and Glazkov because he's beaten Chagaev.

[Lennox]

Massive change in the ratings. So yesterday Boxrec ratings were poo and today they are good?

No Yesterday they were poo and today they are shittier. On what possible objective grounds could Anthony Joshua be number 1?

Because heavyweight is one of the least competitive divisions currently. No heavyweight in the top 20 p4p now - Joshua is #40 ...

the last 2 years
- Joshua defeated Martin at 244 points
- Wilder defeated Stiverne at 181 points
- Ortiz defeated Scott at 135 points
- Pulev defeated Chisora at 101 points
- Povetkin defeated Duhaupas at 101 points

Lennox, you are the greatest

So Charles Martin is a better fighter than Stiverne, Scott, Chisora and Duhaupas? You system is bad because you don't understand boxing and had no correct way to treat the result of Martin - Glakov. Martin ended up with too many points. Charles Martin beat one fighter in he PBO top 100 (ranked 77) before defeating Glazkov. Glazkov got injured, it was a non event.

[BillW]

So Joseoh Parker wins the WBO heavyweight title and drops below two inactive fighters, Glazkov and Martin?
Hell.

[Cap]

As Artie Johnson used to say the ratings are "Verrry Interesting!" Hard to rank boxers who rarely fight and when they do fight, their opponents are third-raters at best. Oh for the good old days....

[JCS]

I'll admit, I haven't studied the latest version of the ratings.. but after all these years, I find it hard to believe that the prediction rate has been improved by so much between this version and the last version.

Having said that, I'm not sure there's any way to declare a "best" version of the ratings.. Using one system's input to compare/contrast with another system's output is like pairing a blind and deaf man together to debate on architectural beauty and using that as a citation for an encyclopedia. If BoxRec were to publish a system that was 100% about predictive accuracy, everyone would be up in arms.. after all, we tried that before about 10 years ago.. the masses were out with their pitchforks.

Also not sure how you can call a fight a "non-event" because a fighter got injured... How can you separate legit injuries from minor or illegitimate ones? That's bordering on bias and just plain making sh|t up.

I also think that some folks responding here don't understand some of the core principles of the system (i.e. inactivity).