Can pool do better than the Accu-Stats TPA?

S

Still_Learning

Shortstop in Training
Silver Member
As many of you know, sometime in the 1980's, Pat Fleming invented a scoring system to measure how well a pool player played over the course of some number of games, usually one match. This is the Accu-Stats "Total Performance Average," or TPA. A player's TPA ranges from zero to one, much like a batting average.

Before I complain about it, I want to clarify that I give all the credit in the world to Pat Fleming for coming up with and promoting it. His TPA is the closest thing I know of to a generally accepted "pool score."

But can't we improve it? Here's what I don't like about the TPA:

1) It's so difficult to use the official scoresheets that I've never met anyone who fills them out. I track matches all the time (informally), and all I ever count is balls and errors. The added value from the complicated scoresheets is beyond me.

2) Scoring has a subjective element that requires non-trivial pool experience. An unplanned safety is a position error, but a planned one is not. Whether a safety was planned or unplanned is not always easy to figure out. And what about two-way shots?

3) The TPA cannot distinguish between meaningless errors and critical ones. All errors are not equal!

4) The TPA works fine for rotation-based games and almost as well for 8-ball, but it doesn't work very well for straight pool, and it's useless for one-pocket.

My dream is a scoring system that (1) is simple to understand and apply, (2) leaves no room for disagreement about how a shot should be counted, (3) works well for all variations of pool, and (4) eventually becomes as well known in sports as the idea of a bowling average or scoring above/below par in golf.

Here's a starting point:

Let's say that all shots are either "good" or "bad." A "good shot" makes you the favorite to win. A "bad shot" makes your opponent the favorite. Your score = number of good shots divided by all shots taken.

The trick, of course, is defining good and bad shots with more precision. Any ideas?
 
Still_Learning said:
...My dream is a scoring system that (1) is simple to understand and apply, (2) leaves no room for disagreement about how a shot should be counted, (3) works well for all variations of pool, and (4) eventually becomes as well known in sports as the idea of a bowling average or scoring above/below par in golf.
Click to expand...

And if you had such a dream scoring system, what would you do with it? Just curious. Since Accu-stats created their system, I used to see it used on TV matches, but it seems to have just gone by the wayside. Rarely, if ever, see those scores any more... So, just wondering, what other useful ideas or usages would there be for it?
 
Still_Learning said:
As many of you know, sometime in the 1980's, Pat Fleming invented a scoring system to measure how well a pool player played over the course of some number of games, usually one match. This is the Accu-Stats "Total Performance Average," or TPA. A player's TPA ranges from zero to one, much like a batting average.

Before I complain about it, I want to clarify that I give all the credit in the world to Pat Fleming for coming up with and promoting it. His TPA is the closest thing I know of to a generally accepted "pool score."

But can't we improve it? Here's what I don't like about the TPA:

1) It's so difficult to use the official scoresheets that I've never met anyone who fills them out. I track matches all the time (informally), and all I ever count is balls and errors. The added value from the complicated scoresheets is beyond me.

2) Scoring has a subjective element that requires non-trivial pool experience. An unplanned safety is a position error, but a planned one is not. Whether a safety was planned or unplanned is not always easy to figure out. And what about two-way shots?

3) The TPA cannot distinguish between meaningless errors and critical ones. All errors are not equal!

4) The TPA works fine for rotation-based games and almost as well for 8-ball, but it doesn't work very well for straight pool, and it's useless for one-pocket.

My dream is a scoring system that (1) is simple to understand and apply, (2) leaves no room for disagreement about how a shot should be counted, (3) works well for all variations of pool, and (4) eventually becomes as well known in sports as the idea of a bowling average or scoring above/below par in golf.

Here's a starting point:

Let's say that all shots are either "good" or "bad." A "good shot" makes you the favorite to win. A "bad shot" makes your opponent the favorite. Your score = number of good shots divided by all shots taken.

The trick, of course, is defining good and bad shots with more precision. Any ideas?
Click to expand...


IMO, Accu-Stats TPA system is by far the best I've seen yet to evaluate a players performance. And I've found it fairly easy to track and score. You basically make a notation for each inning, not each shot.

In all sports the yardsticks that are used to gauge performance have flaws.
A pitcher can pitch great for eight innings and have to leave in the ninth with two outs and the score tied 1-1. The reliever comes in, throws one pitch, and the batter grounds out to end the inning. The first man up in the bottom of the ninth hits a home run to win the game. The reliever gets the Win, not the guy who went eight and two thirds innings and held the opposing team to two hits.

Do you get my drift?
 
Still_Learning said:
The trick, of course, is defining good and bad shots with more precision. Any ideas?
Click to expand...

This is a "new fangled Idea" (actually Mosconi preached it till he died) Make the players "Call the Shots". I know its a radical Idea and the 9 ball bangers get their panties in a bunch if they lose the gambling edge that a good looking slop shot gives them. "scare the opponent make him think you meant to go twice back when you were shooting for once back" I didn't slop the nine in I planned to just miss that corner pocket and send the four half way around the world to barely nudge in that nine"

Make them state what they intend to do if they successfully execute it its a good shot.. if they miss..... they missed.
 
FLICKit said:
And if you had such a dream scoring system, what would you do with it? Just curious. Since Accu-stats created their system, I used to see it used on TV matches, but it seems to have just gone by the wayside. Rarely, if ever, see those scores any more... So, just wondering, what other useful ideas or usages would there be for it?
Click to expand...

Well, to be honest, even if I had a perfect system sitting in my lap, there isn't much *I* could do with it. But if enough people worked to develop one and started talking about it, some momentum might grow out of that.

I recognize it was pretty much sheer fantasy anyway. Pat Fleming's been promoting the heck out of his TPA for 20+ years and probably fewer than 5% of the population that considers themselves pool players has heard of it. And I bet less than 1% knows how to track a match, even the informal way I do it.

I guess pool will have to be like chess...we'll have to settle for relative measures, like who'd win a race to 100, and by how much...

I was hoping one of the APA types might talk about about their rating system. I know they have a messy formula, but it might be worth taking a closer look at.
 
Still_Learning said:
I was hoping one of the APA types might talk about about their rating system. I know they have a messy formula, but it might be worth taking a closer look at.
Click to expand...
Well, actually the APA system isn't too complicated. They focus on keeping track of innings (where each player gets a turn, similar to innings in baseball where you have the top of the inning and the bottom of the inning). So each time both players finish their turn at the table, then they add to the inning count. Thus, if you're a pro player, who breaks and runs most of the time, then that'd be 0 innings. If you don't break and run, but will run out on your successive turn then that's 1 inning. For some people they take 4 or 5 innings. And if you're a relatively beginner player, then they'll take on average more than 5. So, it's a relatively straightforward and easy way to gauge a player's relative strength. Now, of course, there are many other factors that they use in their formula to hone in their skill ratings. For example they make some adjustments for defense, so that it doesn't go against your inning count. And they have ways of determining your rating, even when you have high inning counts in some games, and low inning counts in others...
 
You must log in or register to reply here.
Back
Top