How to fix corrupt scoring in boxing: "altboxrec.com"

This weekend we had another bad scorecard in the Cloud - Campillo fight. Perhaps not as thoroughly outrageous as the Williams -Lara card or the Chisora - Helenius, but nevertheless a "robbery".

When "fixing" scoring comes up, most people try and think of alternative judging systems (i.e. add two judges, put them in a sound proof booth, allow them to use decimals, etc.). These suggestions are futile: boxing promoters love the current system and there is no incentive to change it.

The only way to "fix" the problem, is to ignore the official scorecards and render them irrelevant. In the age of televised boxing and wide availability of full fight videos, the boxing community is fully capable of erecting an alternative and far more accurate verdict than three possibly corrupt and occasionally incompetent judges.

I propose the formation of a website called "altboxrec.com". This website will maintain an alternative history of a boxer's fighting record. It will not duplicate all of boxrec; rather, it will record "corrections" for fights that are deemed by the community to be scored incorrectly.

How would such a thing ever gain credibility? Here's how. When Jim Gray interviewed Gabiel Campillo after the Cloud fight, he stated that he believed that he won, was frustrated, wanted a rematch, etc. But he had no recourse to protest the scoring. With altboxrec.com, Campillo could have said the following: "I reject these scores and await the verdict of altboxrec.com". The moment the website is mentioned by an aggrieved boxer, it has achieved instant credibility.

I've left the hardest part for last. How does "altboxrec.com" determine the winner of a fight? Here's how it would work.

The first step would be for the online boxing community to nominate 50 impartial, experienced boxing scorers. These persons would be nominated on the basis of acquired reputation on boxing forums. It would not include journalists, trainers, promoters, or anyone in the industry. Once the 50 impartial judges have been selected, the process proceeds as follows.

When a controversial decision occurs, altboxrec.com polls it's 50 impartial judges as to whether it should "accept the case". If 50% or more vote 'yea', then the fight is deemed worthy of further review.

Once the "case" is accepted, altboxrec.com randomly selects 20 of its 50 judges. A video of the fight is distributed to each judge for review. Every judge then submits his scorecard to altboxrec.com for further analysis.

The scorecards are then averaged and a margin of error computed (the same way election polls are reported). The resulting average score and its margin of error can be used to decide whether the result should be overturned on altboxrec.com. If so, altboxrec creates a fight entry identical to the corresponding boxrec entry, but with the alternative verdict, plus a link to the boxrec page. The boxers' overall record will also be updated. So, for instance, Tavoris Cloud would have a record of 23-1 and Campillo would be 22-3. (Note, Campillo's record might also benefit from a review of his fight with Beibut Shumenov).

What if some of the 50 nominated impartial judges turned out to be poor or biased? This is easy to determine. If a judges scorecard consistently differs from the average scorecard, it shows that his scores are unreliable (not correlated with the consensus). The first recourse would be to downweight cards submitted by unreliable scorers. The second line of defense would be to eliminate unreliable scorers from the judging pool and to accept nominations for a replacement. In this way the pool of community judges is monitored and maintained.

 

That's not the idea. One person's scorecard cannot overturn a fight. The idea requires a consensus and an estimate of margin of error.

 

The answer is almost even better for close fights like the ones you mention. We'll get 20 scorecards plus a margin of error, and we'll be able to offer precise estmates of the "true score". Most likely close fights such as these will fall within the margin of error -- and guess what: altboxrec.com will call these fights "draws". In other words, there is no consensus winner.

 

Never gonna happen. This requires buy-in from promoters, sanctioning bodies, etc. Forget about changing boxing from the inside. The goal of altboxrec.com is to render the onsite judging irrelevant.

 

Right, by the square root of N. But 20 is a good number, more than that becomes logistically unwieldy.