I feel terrible for yelling at my screen for Matsuyama to block one of those tee shots dead right off 17 and 18...

Man that was fun sweat though.

Will's pitching game was the surprise of the week for me. So nifty around the greens.

As I said on Twitter that last hole was crazy. The tournament was 100% over, but man did it feel like anything could happen with the weight on Hideki. Also just another reason you can't use game theory in golf due to the lack of defense, shared clock, or shared ball. Just play the course and see what happens.

Thanks for all the kind words ITT...it really is an amazing thread to go through!

only 29 players made the cut in both the 2020 US Open and 2021 Masters. only three players have a top 10 in both, and one particular player has the best combined score.

Solid prediction.

One of those in particular doesn't have a PGA Tour card

True but he also has had the largest multiplier on net worth of anyone in golf in 12 months...of course when you only have a couple hundred grand and make $4million it distorts the math!

Amazing...

Scott,

i have a very superficial understanding of your DECADE system and the probability behind it, as well as poker being one of the inspirations behind you developing it. what i was wondering is, if you had to pick, would you compare the probability and strategy behind DECADE more to ring game poker or tournament poker (or have you even thought about it that way)?

the EV of a ring game decision is mostly fixed by the buy-in you play at and the stack size of your opponent, but the same decision in a tournament can be drastically different based on where you're at in the tourney because the prize at the top is such a huge percentage of the total purse. i cut my poker teeth on single table tournaments 15 years ago and the strat at that time was, as an oversimplification, to basically push edges and either bust in 4th or amass a monster stack to cruise to 1st. it seems that Hideki amassed the monster stack (of strokes, not chips) and cruised to the win.

i'm also reminded of Alex Jacob and his Jeopardy! run. AJ was an OG poster in the single table tourney thread and he used a somewhat similar strat to become a 6 time winner and also take down the TOC.

and after typing all of that up and rereading it i just had a serious case of deja vu and think this topic may have been discussed before, either in this thread or another post in this golf forum. haven't gone back to look

I’m not Scott, curious to hear his thoughts. But I’d think it’s akin to LHE ring games. A bunch of tiny +EV decisions that add up to a profit over the course of a tournament.

and that’s the root of the question. should your strat be the same for a random Korn Ferry tournament as it is for a major? ignoring Will’s specific situation where every win was needed for him to gain status, once you are in a free roll situation are the DECADE decisions slightly adjusted to maximize EV? and how is EV measured at the highest level? sure seems like it should be much more than a players share of the purse if total earning is your goal.

obviously I don’t have visibility to the mountain of data Scott and his colleagues have collected, but I do wonder if there are specific shots in unique situations that can result in an expected strokes gained + purse gained outcome that makes it worth pushing the edge.

simplified question: does strokes gained equal purse gained, especially when results can open the door to future opportunity (exemptions) and increase the purse potential?

So DECADE gives the target that shows the lowest average score. But the question is, (if I understand what you're asking) does that target also show the highest birdie percentage?

If I had to guess, I'd guess not always. So if you have to birdie the last hole, then does that change your target? I'd bet he has the data to show highest birdie percentage also.

Golf Digest and/or The Golf Channel would have a field day with this thread as well as a few other gems here.

I assume that on course Will and his caddie have the ability to make decisions on the fly based on score and situation. Like the approach shot on 18 with a 2 shot lead and a 1 shot deficit by necessity have to be different shots. I can't imagine in both of those spots they're just autopiloting based on what their system and planning came up with.

Where'd this happen?

If Ship's twitter's been posted in this thread before I missed it....following now!

Scott, have you ever considered integration with one of the specialized shot tracking apps, eg, Arccos Caddie? Seems like a natural fit (and maybe you already have something like it).

My brother's been using Arccos and it's helped his game a lot. For the price, I'm thinking of getting it.

I see Will is playing in the RBC and is playing with Horschel and DJ.

Yup. Finally found his way into a featured group for round 2.

1:08:57 as the tweet says

Look at me figuring out the multi-quote for the second time! BO is right, it is just finding tons of tiny edges that add up. I started to post it would be like folding a bunch of perfect odds 80/20 against spots to find a better spot to get your money in (that analogy doesn't feel quite right, but sort of right in my head). It's truly just easier to intentionally not lose shots than it is to intentionally gain shots. So you are really tiptoeing around avoiding the -.5 approach shots and amassing a bunch of +.2 shots.

I've said it a million times on Twitter and podcasts, but applying game theory in golf is virtually impossible. The fact you can't impact your opponents play (in a non-******* way like headshots) and there isn't a mutual clock makes it essentially impossible. I asked Will about a few spots and he said "it was just hard to really know what was going on out there because they don't have electronic scoreboards on every hole. Just the big Masters ones a few places." Even Hideki was likely struggling to keep up with what was going on. To add to that, the more energy you spend trying to decipher what the winning score will be the faster your mind starts working and more tired you get. I truly tell my players to just stick to lowest average score decisions and let everyone else distract themselves.

Add to that, IMO, for every one decision you make that strays from optimal scoring average for some game theoretic reason you will probably make 5 bad ones for a net loss. Just play golf.

That all said, Hideki likely could have/should have laid up on 15. Just far too much to go wrong there and very little to gain, he had to know he had a decent lead I would think.

I think those devices make quite a few mistakes and you can't use them in tournaments. I follow the teach a man to fish vs give a man a fish ideology. They also make MANY logical fallacies around the distance you hit each club. Sure it is better than nothing, but not by much IMO. We are working on an update to allow easier on course stats tracking, free GPS range finder feature with targeting help that is more generic and then you have to use the conditions to tighten it in. It should be a great feature for practice rounds for tournament players

It's really pretty comical to see all the hilarious posts and arguments from the past on this site and then to think I was on Wharton's Moneyball podcast!

LOLOLOLOLOLOL!!!!!

obviously i haven't dug into the shots gain stats like you have, but are the majority of the swings as you say? (either -0.5 or +0.2)?

what i'm wondering, based on what you may have found in your data collection, is if there are unique circumstances that can occur which result in the delta between negative and positive strokes gained to be much higher than the typical ring game expectation, therefore making the situation worth considering taking on the additional risk and breaking from the ring game grind (the strat of collecting dozens of tiny edges along the way and not exposing yourself to larger risk swings) to push a potential edge.

not sure if you watched the tourney today or have seen the highlights, but Will's second shot decision on hole 2 today was a questionable one and somewhat of an example of what i'm thinking. took a risk on his second shot, didn't pull it off, and it cost him two strokes. in a vacuum it seemed like a very poor decision at the time, but in certain circumstances it may seem like the better choice.

Will's shot on 2 was a horrific decision. Collin Swatton texted me as he was doing it "this is dumb" right when you heard the ball hit the tree and go OB. That, IMO, was solely because it is a reachable par 5 and he's thinking he needs to get it back up near the green to get a look at birdie. It WAS a birdie hole (and I hate that saying) back on the teebox, it is no longer a birdie hole now.

There are two places you can increase variance a decent amount while not destroying your average score. Off the tee on barely reachable par 4's with hazards too much in play and similar second shots on par fives. You can make quite a few more birdies to offset the increase in bogies. There are spots I would say you can do this & it be correct.

BUT, just being more aggressive on your average 6-9 iron second shot to tuck pins WILL NOT increase your birdie rate while also making more bogies. That kind of aggression is why many guys are stuck on the Korn Ferry. Hilariously, IMO, I have worked with 11 players tied for 12th or higher on the Korn Ferry Tour today. I'd bet my life they ALL said "**** it, I'm committing to DECADE so hard this week. If Will can do it, I can too."

As for the .2 to -.5 approach shots that is a popular range, but it also is a huge range. The best players have the fewest -.5 shots. Now, that is relatively obvious since it is essentially saying "the best players hit the fewest really bad shots" but when I go through ShotLink for my midrange players they have way too many -.4-.6 approach shots with a relatively benign pin from like 130. Those are typically shots where they are trying to force a look with a gap wedge to a pin 3 from the edge instead of targeting 3ish yard toward the middle. Getting line and distance to match up and stuff it is simply a hard thing to intentionally do and most are just variance. Yes the better players have tighter shot patterns, but it's not like they are 50% smaller.

all good info, thanks for sharing.
i suppose what i'm specifically referring to is a situation in the golf tournament as if it were a poker tournament. both tourneys are similar in that they do not have a flat payout structure..they are heavily weighted towards the top. if your goal is to maximize your profit in a poker tournament there are certain situations (i.e. on the bubble) where you consciously take on more risk than normal because the gain in chips you make can vault you to a much better expected value situation moving forward due to the increase in chips and the way you can use them in future hands. Hideki's four stroke lead before the final round seems to be somewhat analogous to that: he was in a position of power and everyone was chasing him, potentially taking on more risk (lower shot gained expectation) in an attempt to catch him.
an example of what i'm wondering: does taking on a slightly greater risk in the third round (say your strokes gained EV is slightly lower on a few shots than your typical strategy with the goal of taking advantage of a situation), but if you are successful your opponents in the final round end up taking on an even greater risk on more shots than they normally would in an attempt to catch you.

total hypothetical example with made up numbers

you: you take three shots lower than optimal strat with a potential loss of 0.3 SG each and total potential gain of 2.0 SG. max delta of 2.9
if you are successful: your opponents end up taking, say, 9 shots with a potential loss of 0.3 SG each and a total potential gain of 2.0 SG in an attempt to catch you. max delta of 4.7