Talk About Tennis 2.0

Watson says 53% Medvedev 47% Djokovic..sounds fair to me ..I would say 55-45, practically the same...

ashkor87 wrote: ↑Sun Sep 10, 2023 5:37 am Watson says 53% Medvedev 47% Djokovic..sounds fair to me ..I would say 55-45, practically the same...

This is a classical recency bias - one victory over Alcaraz erases all previous history between Medvedev and Djokovic (which is very relevant). IMO their H2H 9-5 pretty much reflects their respective chances. It translates to roughly 65-35 in favor of Djokovic.

mick1303 wrote: ↑Sun Sep 10, 2023 5:43 am This is a classical recency bias - one victory over Alcaraz erases all previous history between Medvedev and Djokovic (which is very relevant). IMO their H2H 9-5 pretty much reflects their respective chances. It translates to roughly 65-35 in favor of Djokovic.

I'm not sure that a computer program is likely to suffer from recency bias.

skatingfan wrote: ↑Sun Sep 10, 2023 6:49 am

mick1303 wrote: ↑Sun Sep 10, 2023 5:43 am This is a classical recency bias - one victory over Alcaraz erases all previous history between Medvedev and Djokovic (which is very relevant). IMO their H2H 9-5 pretty much reflects their respective chances. It translates to roughly 65-35 in favor of Djokovic.

I'm not sure that a computer program is likely to suffer from recency bias.

It very much can )) It depends on the weight it puts on the recent matches. I would even say that computer program is more likely to do this thing than a human observer.

Yes, certainly AI can have all kinds of biases but it learns with every match so unless recency is a real factor.. I personally don't think AI adjusts yet for nuances like crowd support, response to pressure, individual motivation or even match-ups of strengths and weaknesses, so I tend to disbelieve the AI prediction, normally. In this case, I was very impressed with M, rather unimpressed with Dj, and expect the crowd to be against both! The court is also fairly fast, which Med likes, So 55-45 Medvedev..

Too many factors to choose from. Hard to prioritize. You can look at success rate specifically in the finals as the measure of ability to handle the biggest pressure there is in tennis. Djokovic won 23 against 12 losses. Almost 2 out of 3. Meds won 1 out of 4 - 25%.

Medvedev was brilliant in the semis. But there was little pressure on him - Alcaraz was very strong betting favorite. This string of predictions tries to put that pressure on him. Although some prominent tennis observers on Yt did pick Djok: namely Gill Gross and Canadian Steven from Slice.

I learned a long time ago not to bet against Novak.
60-40 Novak. But won't be surprised if Daniil wins (as I was when he beat Alcaraz).

A niche tennis reporter at the US Open...this guy works his tail off to find stories about lesser-known players that their local papers will commission. He makes about $50 per story.

https://www.poynter.org/business-work/2 ... -reporter/

Fun story

Berrettini has pulled out with a foot injury, replaced by Zizou Bergs in the draw