Overcoming Mediocrity, NBA-Style



Guest post by Abacus Reveals

Consider, if you will, these two teams:

Team A: Over the past two seasons, these guys converted roughly half (.497) their possessions while allowing the opposition (.499) to do the same.

Team B: Over the past two seasons, these guys won over 100 games, a division championship and a playoff series.

The former seem like a bunch who’d have trouble holding the court for long playing pick-up ball on a Saturday morning; the latter sound like a formidable pro squad, especially if we point out they, like Team A, have been competing in the NBA’s stouter (though maybe not as decisively so this season) Western Conference.

Certainly Team B, unlike Team A, would hold a statistical edge over its opposition in order to have posted such achievement, wouldn’t you think?

Not so fast, though.

You see, these teams not only play in the same conference and division, but also the same city (not Los Angeles) – they are the Portland Trail Blazers.

And yes, they’ve won well over 60 percent of their regular season games whilst being ever so slightly out-played up and down the floor by their opponents.

What makes this oxymoron even more incongruous is that, for all the strategic nuance and micro-analysis in the basketball of today, the winning team in an NBA game “out-converts” (i.e. scores or goes to the line) its opponent closer to 90 percent of the time than one might think.

In 14 of Boston’s first 15 games this season, for example, victory has resulted in a higher conversion percentage, ranging from 3.0 (Toronto loss) to 8.4 (Washington and Brooklyn home blow-outs). The C’s victory over Atlanta was virtually dead even in rate of conversions, 47.4-47.1 in favor of the Hawks.

In only 10 of last season’s 81 playoff contests did the losing team out-convert the winners (two other games produced identical “conversion-to-possession” ratios).  I got a similar 10-15 percent rate of “anomaly” games during the couple of recent seasons when I tracked this odd phenomenon.



So what kind of prestidigitation is being practiced in the Pacific Northwest by Terry Stotts and his apprentice paranormals – and why aren’t they doing it with a team named the Wizards, but I digress!

Well, for one thing, the Blazers play – and win – more of those anomaly games than most. During their 54-win 2013-14 season, they snatched victory from the jaws of out-conversion on 13 separate occasions and won two other contests that graded out equally on the conversion chart.

Only once that season were the tables turned on Terry’s team, in early March by (of all teams) the Lakers. The Portland defense earned two more “stops” (55-53) that night, but LA “erased” three of them by draining eight three pointers while missing only two free throws. The Blazers had nine treys but negated them with an uncharacteristic eight foul line bricks. The result was a one-point Laker triumph. Hmm.

Let’s look at last season’s numbers for that specialty shooting.

In compiling 51 2014-15 victories, the Stotts Squad combined for 317 errant free throws, compared with 807 successful three pointers. The enemy camp had 415 foul-line foibles and 524 drains from distance. That’s a +490 for the Blazers, +109 for their opponents, a 381-point advantage for the season, 4.65 points EVERY GAME.

Think about that … just by playing even with the other guys, you win games by four or five points.
The Trail Blazers’ per-game “spot” was second-best in the Association. If you can’t guess No. 1 and No. 3, then you haven’t been paying attention. (HINT: Houston’s funky offensive philosophy placed them but fourth at +2.45 points per game in a season where so much seemed to go right for them.)




The Celtic Slant

If this simple little computation holds any validity at all, that would be good news for Brad Stevens and his troops. The Celtics finished ’14-15 ranked eighth at +1.32 – in other words, playing even in possessions should produce a narrow victory.

For the second half of last season, this advantage swelled to over three points a game. (Sorry, I don’t have a ranking for the partial seasons.)

Even better, through the first four weeks of this season (13 games through the Brooklyn loss), their “grade” has improved +4.15 – No. 2 in the league to the same no-brainer from before. The Celts’ 2.83-point improvement from last season represents the third biggest jump in the league. (The surprising Knicks are over four points a game better than last year by this standard, Sacramento somehow a smidgeon over three points per game in improvement.)

My pal, TB, always tells me that they track everything in basketball these days, so I’m sure I’ve done nothing more here than reinvent some numerical wheel or other with this simplistic analysis.

[Statistical Footnote: My estimates for possessions were calculated with a formula I began to use five years or so ago – retirement provides the opportunity to scratch some unusual itches, my young friends – so the “conversion” data may not quite jibe with standards hoop reference sources.

My possession totals seem a bit high compared to those calculated by the more complex formula currently offered at basketball-reference.com.  I’m still working at internalizing that alternative formula, and I gotta admit that factor of 1.07 in the middle of the whole thing has me baffled at the moment.]