Curtley
Years ago
Memphis beat Brooklyn today, Ennis had 5 pts, 3 TO, 3PF, 0 Assists and +20 in 28 mins. No other Memphis player was more than +9. Is this a statistical anomaly? Will +/- ever be considered a seriously important stat at the expense of others, the way say scoring, assists, or rebounds are?
Isaac
Years ago
In isolation, he could've just been on for a couple of scoring runs and got lucky. Or maybe his spacing and defence were useful for his team. There would be examples like that in almost every game. Over time, it's useful in looking for patterns amongst roster combinations.
You could ask Andrew Price on Twitter - the Brisbane Bullets have him employed for his advanced stats. I'd be guessing, but I imagine they look more at offensive and defensive ratings than individual stats like points, rebounds and assist (which are probably of more interest to fans). Analysts for NBA teams look at things like opponent percentages when defended by particular players - those things don't show up in points/rebounds/assists - win shares and so on.
Both NBA.com and ESPN include +/- in their boxscores these days. Both have sections for advanced stats and line-up analysis.
Andrew Price's Spatial Jam has some great new stuff up this week for the NBL:
http://www.spatialjam.com/nbl-stat-tracker/
XY
Years ago
Dante Exum +7 so far today whereas the guy he backs up, George Hill is -31.
But Exum has only played 6 minutes for an assist and a rebound in a blowout game. That is why +/- can only be used in context.
Isaac
Years ago
And George Hill is considered a vital piece for Utah (they tend to lose when he's out) while many Jazz fans are going off the boil with Exum.
One combo example Andrew posted on Twitter:
This season, Illawarra Hawks are:To me, that suggests that Bevo is overplaying the Clarke/Martin combo (more possessions than the other pairings). Possibly D from Norton and White being more advantageous for the team.
+97 with Clarke / Norton (>550 pos)
+13 with Clarke / White (>300 pos)
-41 with Clarke / Martin (>600 pos)
Mick
Years ago
The only time I hear raw plus minus used as a stat these days is on this forum when an Australian or ex-NBLer is playing in the NBA and gets a good score. It's not really something that is useful at all to describe a player's impact in an individual game: you need massive sample sizes. Eg. there is little-to-nothing to be gleaned from pointing out James Ennis' +/- from today's game.
Adjusted plus-minus is better, especially when combined with watching the actual game. And even then: these catch all stats are awash with a tonne of noise, and not intended to be used to describe a player's impact in a particular game.
Some people use advance stats like conspiracy theorists. They don't have a clue what is going on in front of them so they clutch on to some obscure factoid in order to make them feel like they know something others don't.
It was abused on here when Delly was in 'beast mode' *cringe*
Roofman
Years ago
Regarding what Mick said above, +- only can be extended to apply to an individual's career (whether real or adjusted) and this tells us nothing we don't already know, because back in the day a player was considered a 'winner' or a stats stuffer on losing teams. People here might or might not know that +- was originally derived from (ice) hockey since it is one of the very few true two way sports and there is greater perspective on it's use there.
You need to be a registered user to post from this location. Register here.
Advertise on Hoops to a very focused, local and sports-keen audience. Email for rates and options.
An Australian basketball forum covering NBL, WNBL, ABL, Juniors plus NBA, WNBA, NZ, Europe, etc | Forum time is: 2:17 pm, Sat 20 Apr 2024 | Posts: 968,026 | Last 7 days: 754
