POST GAME DENVER

Beat,

Hey, I don't claim to be infallible, and my scoring was in pencil with a couple of numbers a bit blurred. But I'll stand by my results until you or someone else comes up with something better.

And the reason is that I believe you completely omitted the extremely important factor of weighting in your quick tabulation. Different players play different numbers of minutes, and therefore each +/- value has to be weighted by the number of minutes played by the player; and then the grand total has to be divided by the sum of the weights (or minutes played by all players combined) in order to get the approximation of the final margin of victory that you were looking for. You may have lucked out on your quick experiment and either (1) all players played about the same number of minutes or (2) deviations in one direction may have been offset by deviations in the other direction.

In the Celtics box score for what I believe is the greatest victory in NBA history:

• The great majority of the minutes played went to four players whose +/- scores were +1, +2, +2, +2....perfectly in keeping with the +2 margin of final victory.

ª Of the other three players, Siegfried and Nelson (at -9 and -4 respectively) were subs who combined to play about the same number of minutes as starter Sam Jones, whose +14 offsets their combined -13.

• So you've got four players, playing the bulk of the minutes, who averaged exactly the +2 final margin of victory. And you've got the other three players whose +/- figures basically cancel each other out.

Everything seems perfectly in order to me. Not even one point is needed from any leprechaun. I'll be extremely interested in your reaction to this explanation.

Sam

P.S. If you try you way on the +/- figures for the Atlanta Hawks last night, I think you'll find that the Hawks should have lost by 26 points. Then we could pile on Doc for not resting his starters. lol.

Sam

Check any box score any one where the plus minus is listed.

Just for an example last nights Atl-Cha game

If you add all the numbers of the Atlanta players you get +125

Atlanta won by 25. Thus with 5 players of ANY combination on the floor the multiply 5 X 25 will equal 125.

This works with any +-.

Take any game any team and it always works.

As for doing this like you did with everything else you kept track of I am amazed. This is a simple math question and the only way I can see a deviation from it is if a team finishes with only 4 players.

Check any boxscore and find one that the winning margin multiplied by 5 does not equal the final +- total of the winning team.

good luck!!

beat

beat wrote:Sam

Check any box score any one where the plus minus is listed.

Just for an example last nights Atl-Cha game

If you add all the numbers of the Atlanta players you get +125

Atlanta won by 25. Thus with 5 players of ANY combination on the floor the multiply 5 X 25 will equal 125.

This works with any +-.

Take any game any team and it always works.

As for doing this like you did with everything else you kept track of I am amazed. This is a simple math question and the only way I can see a deviation from it is if a team finishes with only 4 players.

Check any boxscore and find one that the winning margin multiplied by 5 does not equal the final +- total of the winning team.

good luck!!

beat

beat,

This may very well be so, but I ask "what does it mean?"

If KG played with Dooling, Moore, Daniels and JJJ the whole game, his +/- would probably suck, but he doesn't. He plays with starters most of the time and some sprinkling of early-midbench players.

If starters only played with starters, and only played against starters, then it might be illustrative, but with subs it isn't since they are often mixed and matched. For example, last night Dooling played with subs and almost completely against subs. However, due to our thinness at 5, Steamer played against Zaza, their starter, and with other starters, more than he might under other circumstances.

I have looked at starters' +/- because starters play a lot of time with each other and against the other starters. Sam might disagree with this.

bob

.

Beat, here's your line that I was replying to:

"Question on the above stats if you add all the numbers after the C's you end up with a + 8. Being that those 7 were the only players that played wouldn't the total have to be a +2."

My answer was basically, "No,' because you can't just add all the =/- stats and expect to come out with the margin of difference in the game. That's because the =/- performance of players with more minutes has more influence on the +/- of the entire team than the +- performance of players with fewer minutes does.

Of course, the grand total of the +/- stats, divided by five, will equal the margin of victory. That's a mathematical truism, but it doesn't really prove anything. It's sort of like saying, "125 divided by 5 is equal to 5 times 5." It's true, but what does it prove about basketball?

I will look through my data and see if I might have missed two points. (The numbers beside Siggy's name were particularly difficult to read.) However, it won't change my conclusions about the (lack of) validity of the +/- stat for purposes of evaluating individual performances.

Sam

bobheckler wrote:

beat wrote:Sam

Check any box score any one where the plus minus is listed.

Just for an example last nights Atl-Cha game

If you add all the numbers of the Atlanta players you get +125

Atlanta won by 25. Thus with 5 players of ANY combination on the floor the multiply 5 X 25 will equal 125.

This works with any +-.

Take any game any team and it always works.

As for doing this like you did with everything else you kept track of I am amazed. This is a simple math question and the only way I can see a deviation from it is if a team finishes with only 4 players.

Check any boxscore and find one that the winning margin multiplied by 5 does not equal the final +- total of the winning team.

good luck!!

beat

beat,

This may very well be so, but I ask "what does it mean?"

If KG played with Dooling, Moore, Daniels and JJJ the whole game, his +/- would probably suck, but he doesn't. He plays with starters most of the time and some sprinkling of early-midbench players.

If starters only played with starters, and only played against starters, then it might be illustrative, but with subs it isn't since they are often mixed and matched. For example, last night Dooling played with subs and almost completely against subs. However, due to our thinness at 5, Steamer played against Zaza, their starter, and with other starters, more than he might under other circumstances.

I have looked at starters' +/- because starters play a lot of time with each other and against the other starters. Sam might disagree with this.

bob

.

Where did I say it "meant" anything. I was only trying to figure out the math end of it. And account for Sam's 1969 game 7. Nothing more.

beat

Bob and Sam

esp Sam

I wonder just how many people sit along court side and do the "official" stats. Doing the plus minus would be more than enough for 1 person and Sam I have no clue how you managed to do that game 7 all by yourself with all those hostile fans around you.

and Bob any Celtic five that includes Dooling will always be a minus.

beat

Beat,

I don't know how many people it now takes to keep the +/- stat. I assume they can just use the same running game scores available to all of us and, at the end of the game, hit one button that accesses and applies the +/- program for all players in the game.

Back in '69, the notion of a computer wasn't even a blip on my radar scale. I do recall being pretty much all over the Forum that night—putting up a sign for each Celtics rotation player before the game (they were soon torn down), being at the tunnel to welcome the Celtics (unfortunately, not a lot of support from other Celtics fans in that regard) when they came out at the start of the game and at halftime, and extensively standing and serenading the Los Angelinos when the Celtics mounted an early big lead and had a major run midway in the second half that (as it turned out) put the game out of reach for the Lakers despite Jerry West's heroics.

Also, I wasn't a fan of the Los Angeles program scorecard, which didn't offer space to record my extracurricular stats the way a typical Boston program did. (That's the main reason some of my +/- numbers are hard to read. I had to write them in pencil over printed crap rather than in a nice white space.) I also recorded assists and rebounds for Sam (I had three and seven, respectively, in that game). The only other player I ever did that for was Cousy. For this game, I didn't bother to record my own "bucket chart" showing Sam's shots and scores from each of nine areas on the floor. I had kept that stat throughout most of the 1967-68 and 1968-69 seasons and had finally mtade a presentation of my findings to Sam, who hadn't wanted to look at it because he didn't want to start thinking when he shot. That taught me a lesson on the importance of operating by instinct in the game of pro basketball.

Although I often dump on the ways in which stats are incorrectly (IMO) used in pro basketball, my critiques don't stem from an inherent distaste for stats per se. In fact, I have a huge pile of Red Sox scorecards for the years 1947-1951, in which I kept a lot of different statistics at ages 10-14....especially on the performance of Bobby Doerr, who is still my lifelong idol as well as (for the past 15 years) a really good friend. At nearly 94, he's been the oldest living hall of fame baseball player for a considerable time.

My wife and I visited Bob in his Oregon home about two years ago and took him out to dinner. And I was able to stump him on a statistic about himself. (He loves statistics and has a whole bunch of scrapbooks, kept by his dad and containing the box score from every game Bob played. (The irony was that he never played for stats, being the epitome of a team player. He just loves every phase of the game and even has many baseball movies that he took from the 30s through the 60s.)

Anyway, after having consulted my Red Sox treasure trove before we left on the trip, I was able to recollect for him that I always used to root for the count to get to 1 and 2 when he was at bat. Because it was on that count on which he was considerably more productive than on any other count. He made some crack like, "So you're the reason I always seemed to have two strikes on me."

So I have a deep-seated respect for stats, which continue to have value to me only if I maintain my very critical perception of the ways in which they're used.

Sam

Stats are never wrong, there is just insufficient/incorrect data and bad interpretations