After more controversy over scorecards, boxing sanctioning bodies, pundits and fans are once again debating how best to judge a boxing fight. Here, I test new ideas for scoring systems, and explain how I came to the results published in Boxing News magazine.
When I looked at my own scorecard for Vasyl Lomachenko v Teófimo López back in October 2020 it did not make any sense. For me, López had won the first half of the fight comfortably, outboxing his opponent clearly for six rounds. But Loma sharpened up later, stealing close rounds by a narrow margin, with the rounds counting for the same score under the ‘10-point-must’ system. At the end I had a bizarre draw, 114-114.
Under ‘10-point-must’, a judge awards the winner of a round ten points, with the losing boxer receiving nine, after which any deductions are applied (such as if a boxer is knocked down, or the referee takes further points off for fouling, holding or other misdemeanours).
So, most rounds are scored 10-9 in favour of one boxer, with drawn 10-10 rounds discouraged and rare, and truly dominant 10-8 scores highly unusual in the absence of a knockdown or deduction. Hence, López could dominate half a fight, but Loma could nick the other half narrowly to get a draw. Every round was 10-9, regardless of the winning margin.
This was just my unofficial scorecard, scribbled for fun at home on the sofa. The real judges in Las Vegas had seen the bout in favour of López the winner, and we were spared controversy. But this was also the night of Lewis Ritson v Miguel Vázquez, a fight whose aftermath was more concerned with criticising the scoring than reviewing the fight itself. Then came Zelfa Barrett v Kiko Martínez, and a frenzy of podcasts, articles and editor’s letters debating how to score boxing fights fairly.
The ‘10-point-must’ system allows a boxer like Loma to stay in touch despite being comfortably beaten for most of a bout, and permits a boxer to open a wide and unassailable winning margin even in a tight contest where they are only marginally edging debatable rounds. But ‘10-point-must’ only became standard in the middle of the twentieth century, succeeding all sorts of other scoring methods: referees awarding a single point each round, ‘5-point-must’, and even bouts where the ringside reporters agreed the winner and published the decision in the papers the following day. ‘10-point-must’ is not set in stone, and recently, many commentators, pundits and fans have suggested ideas for reform.
But what would the results of contentious fights actually have been if scored in a different way?
Last week I spent a morning building a spreadsheet model that allows us to re-run real fights and compare the outcomes under any number of new hypothetical scoring systems.
I used the model to test eight variations, analysing the results generated by each for four recent contested decisions: Deontay Wilder v Tyson Fury 1, Canelo v GGG 1 and 2, and Josh Taylor v José Ramírez. Each fight was close, and is still debated by journalists, commentators and fans.
My methods, results and analysis were published in Boxing News magazine in August 2021.
Here, I explain how I devised and tested the model, and you can download the spreadsheet to test and amend it yourself here:
Ideas for change:
Finding suggestions for new scoring systems was easy.
After Barrett v Martínez, the Chairman of the British Boxing Board of Control, Robert Smith, defended his judges, but suggested to Boxing News that ‘10-point-must’ might be improved to reward a clear win more generously than edging a round. The magazine’s editor, Matt Christie, proposed using more latitude in the ten-point system – a close round scored 10-9, a clear one 10-8, all the way to 10-5 for a “shellacking” – and readers had their say, with Roy Brand proposing the fighters share ten points, 5-5 for an even round, 6-4 being close, all the way to 10-0 for absolute domination with two or more knockdowns.
Christie also suggested scoring more rounds level 10-10 might reduce “misleading totals” – wide winning margins in contests that are actually close and competitive. But many, such as DAZN’s Mike Costello, feel this would produce too many narrow, disputed scores. For me, the argument misunderstands the role of the scoring system itself. Judging is subjective, and if a fight is close the score should be too. The question is, are the right fights close and contentious, and what does ‘right’ mean?
And a fortnight on from more controversy – this time surrounding the bout between Jermell Charlo and Brian Castaño – the World Boxing Council (the sanctioning body commonly known as the WBC) went public with its own suggestion for change. Their ‘Quantitative and Qualitative’ – or ‘Q&Q’ – system seeks to address the confusingly wide or artificially narrow scores created by ‘10-point-must’.
The model:
I programmed five systems at first:
the familiar ‘10-point-must’
the old ‘single-point-round’
the WBC’s new ‘Q&Q’, where rounds are awarded at one of four levels: ‘Close’, ‘Moderate’ and ‘Decisive’, all scored 10-9, or ‘Extremely Decisive’, where one boxer is ‘significantly outperformed…dominated and staggered’, scored 10-8
Roy Brand’s ‘Share-of-10’ where the boxers share ten points each round: 5-5 for a draw, 6-4 for a debatable round, 7-3 competitive but clear, 8-2 clear, 9-1 dominant with a knockdown, and 10-0 totally dominant with two or more knockdowns, and
Matt Christie’s debatable draws, or ‘DD’, where close and contestable rounds are awarded 10-10, rather than a forced 10-9 as in ‘10-point-must’.
For fun, I also tested several round-by-round systems that turn three judges’ assessments into one score. These were:
‘Unanimous’ (one point awarded only if all three judges agree the round winner)
‘Majority’ (where the point is awarded if at least two judges agree)
‘U8/M9’ (where the round is scored 10-8 if all three judges agree the winner, 10-9 if a majority go for one boxer, and 10-10 if the judges are split three ways including a draw)
‘Judges’ simply use a menu to select the winner of each round and indicate how clear the win is. The model then computes the scorecards under each system and summarises the results.
To check the model worked, I entered my card for Loma v López. Each new system gave the win to López; hardly surprising when I had been seeking to correct my draw. But analysis of the official scorecards for recent close and contentious fights – Wilder v Fury 1, Canelo v GGG 1 and 2, and Taylor v Ramírez – produced the interesting results detailed in Boxing News.
A note: It is impossible to know how the judges saw each round beyond knowing their chosen winner. I watched the fights using the combined judges’ scores, online polls of fan opinion, press reaction to the bouts, and my best common sense to make the call each round.
In Beta: testing the model
To investigate the implications of the new systems with known judgments, I joined up with two co-judges – my brother Harry and friend Tom – and put the model to the test on the three contentious fights; Lomachenko v López, Ritson v Vázquez, and Barrett v Martínez.
The rules were simple: turn the sound off to eliminate the influence of the crowd and commentary, no watching between rounds to see how the broadcasters’ experts had it, and no peeking at each other’s scores.
Vasyl Lomachenko v Teófimo López – 17 October 2020
López Won by Unanimous Decision:
Tim Cheatham 112-116 | Julie Lederman 109-119 | Steve Weisfeld 111-117
The data for Loma v López were the simplest to analyse.
López got a majority win under ‘10-point-must’, with my draw skewing things. But when I was able to account for the winning margin in each round, the ‘Q&Q’ and ‘Share-of-10’ systems each gave it unanimously to López. These systems had smoothed out an artificially close score forced by the requirement to choose a winner by a 10-9 margin, and López was the champion; hardly surprising when the test was partly a response to the draw with which I was uneasy.
Lewis Ritson v Miguel Vazquez – 17 October 2020
Ritson Won by Split Decision:
Michael Alexander 115-113 | Marcus McDonnell 113-116 | Terry O'Connor 117-111
Ritson v Vázquez was simple for different reasons.
The judges had scored the bout a split-decision win for Ritson, and Terry O’Connor’s score of 117-111, a margin of six rounds to three, drew derision. No matter how we cooked it, every judge saw the fight just as Matt Macklin had for Sky Sports on the night: a clear win for Vázquez. The margins of victory varied slightly depending on how dominant the judge felt the winner had been, but there was no argument. Put simply, Terry O’Connor had seen the fight differently to us (and most viewers). Multiple judges are used for a reason: being ringside, viewing from different angles, no cameras, and the influence of the crowd all play a part in how a judge perceives a contest. No sensible scoring system would turn O’Connor’s score into a win for Vázquez, and nor should it. My model was designed to test scoring systems, not to question officiating.
Zelfa Barrett v Kiko Martinez – 13 Feb 2021
Barrett Won by Unanimous Decision:
Howard Foster 116-113 | Steve Gray 118-111 | Bob Williams 118-111
Barrett v Martínez was messy and enlightening.
There were several close rounds where one judge awarded a debatable decision one way or the other. There was consensus that the later rounds that went to Barrett were close, and that Martínez had won some clearly, particularly early in the fight. Between us, Barrett took the fight on ‘10-point-must’ via unanimous decision, with scores of 116-113, 116-114 and 115-114, and by the slimmest of margins on the ‘Share-of-10’ where just two points separated the fighters; the equivalent of scoring a round as “close but secure” rather than “debatable”. But under ‘Q&Q’ we get a split draw at 116-113, 114-114, and 112-113, a result that might appeal to Andy Clarke who scored the contest a draw on the Sky broadcast. Martínez had taken fewer rounds, but won them clearly, and this had made up for Barrett winning more rounds but by a debated margin.
What to do?
We accept that boxing bouts can be close and debatable; the score should reflect that, and debate will ensue. But perhaps that gives us all something to talk about, enriches boxing’s history, and provides fuel for promoters to grow the sport. And we must avoid a crisis in confidence associated with its subjectivity that could confuse fans and turn people away.
Plus, a note of caution from football: Three points for a win in a league match was first trialled in 1981, and became standard worldwide as late as 1995. Originally proposed to incentivise goal-scoring, it had the opposite effect, making the game more defensive as teams feared losing more than they valued winning. There are now fewer goals and more fouls. Tinkering with the rules can have unintended consequences, and other sports are not immune to the problem of subjectivity.
The scientist in me believes the answer to finding a new and fairer scoring system – if such a thing exists – may lie in conducting a wide-ranging test in the image of mine, running professional judges’ assessments through several systems on the night and analysing the outcomes. Allowing a few level rounds might avoid skewing results one way or the other, or perhaps the WBC or Roy Brand will put their name to a system adopted across the world as the new standard.
Late on Sunday evening, myself, Harry and Tom couldn’t resist the urge to revisit one of the most contentious of all decisions, Hagler v Leonard from April 1987. On the night, Leonard won a controversial split decision, 113-115, 118-110, and 115-113. Hagler famously never came to terms with defeat, and the dispute between fans and commentators continues. But in silence, at home on a Sunday, we all saw the fight the same way; a clear and comprehensive win for Leonard under all systems in the model.
Boxing scoring is subjective, many fights are close and contentious, and much of the controversy is inevitable. Some of it, though, we make ourselves.
*
Marvelous Marvin Hagler v 'Sugar' Ray Leonard – 6 April 1987
Leonard Won by Split Decision:
Lou Filippo 113-115 | Jose Juan Guerra 118-110 | Dave Moretti 115-113
To subscribe and read my analysis in Boxing News: https://www.boxingnewsonline.net/subscriptions/
*
As well as Lomachenko v López, the fights I re-ran using the model and real judges scorecards were:
Tyson Fury v Deontay Wilder 1 - 1 December 2018
Draw by Split Decision:
Phil Edwards 113-113 | Alejandro Rochin 115-111 | Robert Tapper 112-114
Canelo v GGG 1 - 16 September 2017
Draw by Split Decision:
Adalaide Byrd 110-118 | Dave Moretti 115-113 | Don Trella 114-114
Canelo v GGG 2 - 15 September 2018
Canelo Won by Split Decision:
Glenn Feldman 114-114 | Dave Moretti 115-113 | Steve Weisfeld 115-113
Josh Taylor v José Ramirez - 22 May 2021
Taylor Won by Unanimous Decision:
Tim Cheatham 114-112 | Dave Moretti 114-112 | Steve Weisfeld 114-112
*
Download my spreadsheet to test other fights and enter your own scores HERE:
I have left my cards in, and made three blank fight tabs. You can copy the master tab to create new blank cards to use.
*
