Statistics, AI, and the Champion’s Mind

 

Watching the 2022 Australian Open Men's final match between Rafael Nadal and Daniil Medvedev was a treat to any sports lover. Or for that matter for anyone who likes exciting entertainment. That it was a duel between two high ranking tennis professionals with great skill and grit is just understating facts. So many comments and commentaries have already been made about this exciting match with a nail-biting finish at the Rod Laver Arena in Melbourne on the 30^th January 2022. This piece is not about the match; rather this is about how our minds get conditioned and confused with statistics and collective intelligence or lack of it.

Behind the fascinating game is a curious play of statistical probability that has played differently in the minds of people and at different times of the match. Well, before the starting of the match vast majority of spectators have gone with Nadal as the winner looking at his track record and the fact that there was no Djokovic added to their belief. Surely there were some people who have placed their bets on Medvedev looking at the way he has fought big battles on the world tennis circuit in the recent years. Moreover, Medvedev was seen as younger, taller and a strong underdog while Nadal was seen as coming back after long injury break and a good 10 year older than his youthful opponent at the finals.

The mind map of spectators changed a bit after Nadal lost the first set without much of a fight. Some doubts would have creeped into the minds of some people but still most of them kept their faith in Nadal. In the minds of most spectators the pendulum almost fully swung into Medvedev’s court after his straight second set win, though this set was harder for both. In a match that needs three out of five sets to win one way to look at the statistics is that Medvedev’s probability has already crossed the 66.6%, ie., two out of three. For the moment, only very few people would look at even giving Nadal the remaining 33.3% probability of winning the remaining three sets. Because that is a difficult proposition, and the mind looks for easier options based on ‘historical data’ which is coloured often by the ‘recency effect’ of the latest set of data. For most of us future is just a scoping of the past and present to a longer timeframe with minor changes.

The confirmation bias is almost completely tilted in favour of Medvedev when the score board showed 0-40 (offering three breakpoints to Medvedev) in the very first game of the third. At this point the tournament’s AI was giving Medvedev 96% chance of winning. Afterall, what is the need to suspect the AI? Unlike human brains the AI has no bias!

Really? No. AI too is biased by a program that works on statistical algorithms created by humans. Or that is how it works mostly as of now.

Well, after Nadal won the third set people started suspecting the AI. And of course, the AI too would have started revising the calculations, just like the Google Maps re-routing the way after a wrong turn taken by the driver. And then goes the fourth set to Nadal and statistically the odds are even now! But the spectators’ minds have almost tilted towards Nadal as that is the outcome most of them wanted anyway. They wanted to go with a champion who is going to create a historic record and not with an upstart young player, however powerful he may be.

Now the question is what would have been Nadal’s state of mind during this dramatic match at various stages swinging from deep despair to high hopes? We wouldn’t know that. I would imagine that with a champion’s mind Nadal would never have given up hope. In fact, during the entire match there was no significant reaction from Nadal to indicate that he has given up on the match. On the contrary, he kept playing despite all the odds and probably an aching body.

At the end of the long match, Nadal could barely walk to the presentation podium. In fact, we saw him sitting down during the presentation ceremony out of sheer fatigue and pain on his legs. A great champion’s mind cannot be constrained by statistics and AI. True champions will junk all statistics if they don’t help boost their winning chances.

Well, all this is based on what we saw on the final day of the tournament. There could be a lot of ‘what if’ scenarios that would have changed the outcome and the analysis. We do not know whether the tournament would have gone this way if Djokovic was part of the tournament. What if by sheer stroke of a couple of missteps by Nadal or smart flukes by Medvedev the fifth set would have gone in favour of Medvedev? All these could be possibilities.

Statistics normally deals with larger number of events or historical data from which we derive certain probability calculations. These probability calculations keep changing as the data set changes. Yet, the probability calculations can go wrong in real life situations. Why? Any Probability number that is less than 100% always leaves a chance of going wrong. For instance, 99.9% probability of winning still leaves a chance of losing. Luck often plays at the margin. That brings us to the realm of possibilities. What is possible may not have occurred in the past at all, yet it could just emerge in the future. And the future could be the next moment or after a million years.  

Nadal did not allow statistics to come in the way of his determination. For him what he imagined appeared more real than what appeared on the score board during the game.

‘Everything you can imagine is real.’- Pablo Picasso