I have a slight problem with TS percentage now.

[Carbine]

I do not see the logical reasoning for 1/2 free throws to be worth a 57 TS percentage.

If I take two lay ups and make one of them, my TS is 50 and I'm going to hear about inefficient scoring

If I get fouled twice on lay ups and make 2/4 FT's, I'm an efficient scorer at 57 TS - even though I am no better than the first example.

Isn't the logical thing to make 5/10 FTs worth 50 TS percent? The fact I can go 10/10 on free throws and have a TS percentage above 100 (113.5) is absurd to me.

[Roundball_Rock]

I do not see the logical reasoning for 1/2 free throws to be worth a 57 TS percentage.

It's because some FTA attempts come from 3 point attempts and you get an opportunity to score 3 points, not 2. Any formula will have some flaws but I get what they were trying to in trying to factor in that not all FTA's are the same but they had to assign a somewhat arbitrary number in an attempt to capture that. They couldn't assign it a value of 0.5 or 0.33 because of the discrepancy based on whether it was from a 2 point or 3 point attempt and settled on 0.44 as the coefficient.

Some of this stuff is overblown. As you noted in another thread, 1 extra shot notably boosts the top line TS %. For example:

Player A: 20 points on 20 FGA, 0 FTA=50%
Player B: 20 points on 19 FGA, 0 FTA=53%
Player C: 15 points on 15 FGA, 0 FTA=50%
Player D: 15 points on 14 FGA, 0 FTA=54%

If a guy takes some end of shot clock or end of quarter bailouts and his TS % splits decrease 3-4% as a result, is that the big deal it is treated as by fans when we are talking 1 extra missed shot in 48 minutes?

[Shogon]

Stats are a guide and will point you in the right direction but you have to watch the ****ing games.

One day, when the stats are truly advanced enough, you won't have to watch the games, but that is unlikely in our lifetimes... maybe not with a miracle AI but I digress... at which point basketball will be a thing of the past anyways.

[TheCorporation]

I never did like TS to be honest

Give me FG, 3PT, and FT separately as they are just that. Seperate measures of efficiency

[tontoz]

I do not see the logical reasoning for 1/2 free throws to be worth a 57 TS percentage.

If I take two lay ups and make one of them, my TS is 50 and I'm going to hear about inefficient scoring

If I get fouled twice on lay ups and make 2/4 FT's, I'm an efficient scorer at 57 TS - even though I am no better than the first example.

Isn't the logical thing to make 5/10 FTs worth 50 TS percent? The fact I can go 10/10 on free throws and have a TS percentage above 100 (113.5) is absurd to me.

As RR mentioned it isn't that simple. How do you account for an and1 free throw? Technicals?

If you assign half a possession to each free throw than several trips down the court will suddenly become 1.5 possessions instead of 1 so they had to make an adjustment.

[Bronbron23]

I do not see the logical reasoning for 1/2 free throws to be worth a 57 TS percentage.

If I take two lay ups and make one of them, my TS is 50 and I'm going to hear about inefficient scoring

If I get fouled twice on lay ups and make 2/4 FT's, I'm an efficient scorer at 57 TS - even though I am no better than the first example.

Isn't the logical thing to make 5/10 FTs worth 50 TS percent? The fact I can go 10/10 on free throws and have a TS percentage above 100 (113.5) is absurd to me.

Been saying this for a while now. It makes ts trash tbh.

[tpols]

You can get points from FTs without even using a possession.

[TheCorporation]

Take 2, scenarios, two-possesions each then:

Scenario A: shoots 4 ft, makes 2
Scenario B: shoots 2 fg, makes 1

What is TS now? 2pts each on 2 possesions

[tpols]

Also, for And-1's... by definition it was already a successful possession. Whether you hit the FT or not is just icing on the cake.

[Carbine]

As RR mentioned it isn't that simple. How do you account for an and1 free throw? Technicals?

If you assign half a possession to each free throw than several trips down the court will suddenly become 1.5 possessions instead of 1 so they had to make an adjustment.

This scenario brought another flaw in it, but on the flip side.

If I get an And1 and convert it, the TS should be the same as if I hit a 3 pointer. Logically speaking - one possession and my team gained 3 points.

As it stands now, an AND1 with the converted FT is 104 TS and a 3 ball is worth 150 TS....... in reality they're worth the same, but the 3 ball makes you 50 percent more efficient?

That scenario is borderline insane to me. In reality I'd rather the And-1 with the benefit of adding foul totals to the other team.

[Roundball_Rock]

Take 2, scenarios, two-possesions each then:

Scenario A: shoots 4 ft, makes 2
Scenario B: shoots 2 fg, makes 1

What is TS now? 2pts each on 2 possesions

A: 57%
B: 50% (assuming it was a two pointer)

[StrongLurk]

I simply use EFG% and FT% separately. Volume of shots helps too.

[bladefd]

I never did like TS to be honest

Give me FG, 3PT, and FT separately as they are just that. Seperate measures of efficiency

I do the same thing. I don't even understand the point of TS. A competent GM would break it down into separate components and judge accordingly. I don't like TS because it tries to weigh different things combined together into one single statistic. You can't tell very much from TS alone without seeing all of the pieces that make it up..

[FromDowntown]

A: 57%
B: 50% (assuming it was a two pointer)

Then that would be the problem. Odd.

Are they saying that 50% from the line is worth more than 50% from the field? (assuming all FG taken were 2-pointers)

[Kblaze8855]

There is no reason for any of these numbers that combine existing numbers to exist for one very good reason....


Every single person dedicated enough to be looking at some athletes “advanced” stats is gonna look at the whole stat line when trying to evaluate them anyway.

The kind of obsessed nerds who have these arguments are all over stat pages seeing the very information the formulas combine and the kinds of people who aren’t gonna consider advanced stats don’t care when you offer them anyway. So who gains?

Someone tells you ____ has a TS% of whatever vs ____ who has a different one....if you don’t already know their games to know why...you just go look at the different numbers to know why. If you need that step to have the required information why have the combined number to begin with?

[tontoz]

I simply use EFG% and FT% separately. Volume of shots helps too.


EFG% is fine but for whatever reason doesn't get brought up much. People bring up FG% all the time even though it is useless for many players.

The reason TS is useful is that it combines percentages and volume. People are posting these individual scenarios that look bad on small volume but over a large sample they normalize.

If two guys are shooting 80% from the foul line but one guy takes twice as many then he is more efficient. TS takes that into account.

[NBAGOAT]

they just track points per possession now but ts% is easily available for us. the .44 multiplier is fine however

[Roundball_Rock]

they just track points per possession now

That stat has its own flaws--just look at the players who wind up being the leaders in it. Are these the 10 best offensive players?

1. Ivica Zubac • LAC 133.5
2. Jarrett Allen • BRK 132.4
3. George Hill • MIL 130.7
4. Rudy Gobert • UTA 130.3
5. Daniel Theis • BOS 130.3
6. Brandon Clarke • MEM 127.0
7. Thomas Bryant • WAS 126.6
8. Damian Lillard • POR 126.2
9. Hassan Whiteside • POR 125.2
10. Clint Capela • HOU 124.0

The oRTG formula favors low volume players who catch and dunk (if big men) or catch and shoot (if perimeter).

Offensive BPM gives you a reasonable list:


1. Damian Lillard • POR 8.3
2. James Harden • HOU 8.2
3. Giannis Antetokounmpo • MIL 7.5
4. Luka Dončić • DAL 7.4
5. LeBron James • LAL 6.7
6. Kawhi Leonard • LAC 6.5
7. Trae Young • ATL 6.3
8. Anthony Davis • LAL 5.5
9. Nikola Jokić • DEN 5.4
10. Bradley Beal • WAS

[NBAGOAT]

That stat has its own flaws--just look at the players who wind up being the leaders in it. Are these the 10 best offensive players?

1. Ivica Zubac • LAC 133.5
2. Jarrett Allen • BRK 132.4
3. George Hill • MIL 130.7
4. Rudy Gobert • UTA 130.3
5. Daniel Theis • BOS 130.3
6. Brandon Clarke • MEM 127.0
7. Thomas Bryant • WAS 126.6
8. Damian Lillard • POR 126.2
9. Hassan Whiteside • POR 125.2
10. Clint Capela • HOU 124.0

The oRTG formula favors low volume players who catch and dunk (if big men) or catch and shoot (if perimeter).

Offensive BPM gives you a reasonable list:


1. Damian Lillard • POR 8.3
2. James Harden • HOU 8.2
3. Giannis Antetokounmpo • MIL 7.5
4. Luka Dončić • DAL 7.4
5. LeBron James • LAL 6.7
6. Kawhi Leonard • LAC 6.5
7. Trae Young • ATL 6.3
8. Anthony Davis • LAL 5.5
9. Nikola Jokić • DEN 5.4
10. Bradley Beal • WAS

well ts% was never about finding the best players either, just the most efficient. and ortg is not ppp which isnt on bball reference. I think second spectrum just tracks every play and counts it as a possession if the guy scores, gets to the ft line or commits a to(one advantage of ppp accounts for tos). also you're not using .44 multiplier for everyone but specifically tracking how much each guy scores

[insidehoops]

I've never looked into this one particular issue, but a guess would be that if you score 2 points via a FG then that's "all" you did, but if you score two points via two free throws then you scored two points AND you drew a foul on an opposing player, and drawing fouls on players has value since it may limit their minutes on the floor and force the opposing coach to sub a lesser player in instead.

Just a 100% guess there, though. Not an official opinion or anything. Absolutely no idea if that's the thought process.

[Carbine]

It's not.

[GimmeThat]

well, by trying to make 2/4 FT in 50% TS instead of 57% would then be trying to normalize the FT line as the standard distance for 2 pt FG. the current 57% implies that the equation has accounted for the fact that players take FGA closer to the basket than the FT line.

one may try to argue off ball fouls etc. balances out corner 3's due to its distance on getting back on defense.

I see that the reason of pushing the 50% TS for 2/4 FT being, there's a preference in keeping the ball high up instead of down low when making contact in the post, the science behind it being it avoids back injury. as for the classic up and under, it's a move that actually requires players to be able to get the ball up high, and score without making any dribble.

[ncrizzle]

very interesting thread