Clutch Shooting Percentage (CSP): A New Cue Sports Stat (sabermetric)

http://poolst.at/csp

We are happy to announce first here on AzBilliards, after many hours of mathematical analysis, our formula for a new stat, considered a sabermetric, in the cue sports world: Clutch Shooting Percentage (CSP). This stat measures the effectiveness for a player under adversity, i.e., their ability to perform in a game/match when it matters most.

A similar metric exists in basketball, however, it is a completely different formula. Ours is a function/recurrence relation in three variables: x,y,z. We weight the equation with λ and α accordingly to 8,9 & 10 ball games depending on which calculation is necessary for the measure of a players ability to perform effectively. While we are not releasing the full disclosure of the variables or weights at this time, we believe we have an accurate representation that provides a suitable measure of clutch play in these game formats.

In short, we consider opponent balls potted, consecutive balls potted, and player balls potted, weighting each for a precision sabermetric never before seen in the cue sports. These weights will adjust overtime as we analyze more data and can make better predictions of a player’s clutch performance. There is also an extra calculation in the algorithm within the Pool Stats Pro app, that applies to missed shots. That is just a factor and is not needed in the general formula. A maximum value for CSP during a game is 100.

Check out the full press release and formula at:

https://poolstats.co/clutch-shooting-percentage-csp/

Rankings that we have so far with total data points are at the first link in the above. Feel free to ask me any questions regarding this. Please be patient in waiting for a response.

http://poolst.at/pdf2

ABSTRACT

The aim of this paper is to describe and list a new Pool Stats metric, which we call a sabermetric, after the notion that was introduced to the sport of baseball by Bill James. We define 'clutch' to mean a player's performance under adversity and executing the shots needed to seal a victory. Similarly, a player's match 'clutch' performance is the average performance throughout each game. That is to say, if the player performs stronger toward the end of the match the overall CSP will reflect that. We will outline our formula for CSP and CSP-Match (CSPM) and give some brief calculations and examples to give the reader an idea of the values CSP produces. The values of CSP are calculated within the Pool Stats Pro app and are available in the app and on the Pool Stats Metabase interface.

Cron said:

Another mathematical mystery... hard pass.

It's 2020 friend, nobody is going to accept that a simple formula should be closed. The only reason Fargo is getting away with it is because American Pool isn't currently popular enough for enough people to care.

Click to expand...

If you would've looked at the white paper you would've seen we disclosed everything. We kept it closed until the release of the formal White Paper for the sake of brevity.

garczar said:

Is it April 1st?? Is this for real??

Click to expand...

It's May 23rd and this is very real.

slach said:

Are these stats all calculated by just entering the result (made, missed, safety, fouled) of each shot during a game?

Click to expand...

The CSP values are calculated only after the last consecutive shot by a player has been made or if the player misses. That is if they hit the 1 on the break, pocket the 2, 3, and 4 but play safety on the 5 ball, then CSP is calculated after that last shot and it is automatically calculated using the app, which then gets uploaded to our server.

Defense is not considered in this equation, although important, it is not a heavy factor in clutch shooting. Yes, it is important, but with the data we have seen, addding defense to the equation will only alter it slightly. A typical professional pool player will play defense on the rate of 5-9 shots per 100 shots taken. When calculating the DSR on that small of sample size during a match, it will only alter CSP slightly (2-4% alteration of the overall CSP value). Instead, we have a stat called DSR (Defensive Success Rate) http://poolst.at/dsr which will measure a player's effectiveness during safety play through the course of a year.

slach said:

Is the difficulty of the shot considered? If so who decides how difficult the shot was?

Click to expand...

Yes, the probability of the shot is considered. A person who is well informed in the game of pool decides how difficult the shot was. It is a subjective opinion, as some shots may be easier for some players than other, but we try to generalize the degree of difficulty for each shot.

slach said:

Who decides if there was a miss or if it was a safety.

Click to expand...

Intention is a hard thing to gauge at any level. We try to make our best decision as regarding a player's intention. Not every data point is going to be 100% accurate, and we approach this with aiming for minimal error. This is statistics afterall.

slach said:

Who enters all this stuff and makes these subjective decisions - player or an observer?

Click to expand...

A player can enter each shot for every player they play. That's an option. A better option is to have a knowledgeable 3rd party. You think when they record baseball, basketball, etc. stats that every data point is 100% accurate? They have a team of knowledgeable data entry personal to do the dirty work. So do we.

If you are considering downloading the app to record your data, you will get familar with how to record quickly and with least amount of error. There is also an undo button in case you mess up. My colleague and I play pool often (though not for the past few months obviously), and I record our data for every match, while I play the game. It adds about 4 seconds to my shot, but I'm okay with it.

And you're right, it's subjective. But so is everything. The aim in recording data and in life is to be as objective as possible; but, nothing is fully objective to an individual.

Geosnooker said:

Akin to resolving the disparity of General Relativity and Quantum Mechanics.

The great snooker player and philosopher Ronnie O’Sullivan best summed up the secrets of billiards after winning his third world championship and being asked about his approach.

‘I don’t know. I just try and shoot the ball into the pocket’

Click to expand...

That's right. My undergraduate and graduate school statistics professors both said the same thing. "Statistics basically says, 'Welp! That's good enough.'"

And for the most part they have been shown to be 'good enough' .

Hits 'em Hard said:

Your logic is miles off. If a 5-10% portion of a players game doesn’t have a 5-10% impact on whatever performance metric you want to use. It’s flawed. There have been matches won and lost on safeties. This entire ‘formula’ is craptastic. Please go back to the pool table and learn how to play pool before you try and influence pool.

Click to expand...

5-10% of the shots are defensive; however, not every defensive shot leads to victory, only the defensive shots that get you the ball back, ergo it wouldn't be the full 5-10% that effects the game, but the percentage of defensive shots that get you the ball back. If you make one bad defensive shot and your opponent misses on their next turn, than that defensive shot has no bearing on your game.

That's why there is the term 'containing safety'.

Also, I said it affects the value of CSP by 2-4% because it's a function with multiple variables and every variable has a different effect on the overall value of CSP, because it's non-linear - like almost everything in nature.

Think of it another way. If you spend 5-10% of your time on a project at work, do you get 5-10% of the project done? Not necessarily, unless the project is linear, which is improbable. You might only get 2% of the projected completed, or maybe 15% if it's easy.

Forthwith, I will only explain the math to people who have read the white paper. My job is not to educate people about basic math.

Hits 'em Hard said:

That’s not how stats work there bud. If I spend 10% of an 8 hour shift on a 50 hour project, then I can make your stats work. But if it’s kept in the realm of one work day, like one game of pool. Then everything you’re trying to explain breaks. You’re not as smart as you think. Go take a 300 level Prob and Stats class at college. You might gain an understanding.

Click to expand...

::sigh:: You're still off. You can finish a 2 hour project in 1 hour if you are efficient. Likewise, you can spend 10 hours on a 6 hour project because of unknown variables and you may have done things during that 8 hour day that you should have finished a 6 hour project in, all because you did things that didn't contribute to the success of the project. Such as playing a containing safety, which is a neutral effect on the game. The game would go on forever with no end if each player played a containing safety everytime.

Also, you are looking at everything in linear terms. Life is non-linear my friend. And so is pool.

I plan on staying around, but how is one to deal with a person who passes stern judgement on a paper they didn't even read?

I took an 800 level Probability & Statistics couse in graduate school, where my M.S. was in Mathematics with a focus on Number Theory. I minored in Stats in undergrad.