NERD Game Scores: A Small Change in the Methodology

Devised originally in response to a challenge issued by viscount of the internet Rob Neyer, and expanded at the request of nobody, NERD scores represent an attempt to summarize in one number (and on a scale of 0-10) the likely aesthetic appeal or watchability, for the learned fan, of a player or team or game. Read more about the components of and formulae for NERD scores here.

***

Most Highly Rated Game
Los Angeles NL at New York NL | 19:10 ET
Kershaw (131.0 IP, 52 xFIP-) vs. Colon (109.1 IP, 99 xFIP-)
One overlooked advantage to having invented a hastily constructed algorithm for estimating the likely aesthetic appeal of baseball games is that one is able to alter said algorithm at his or her own pleasure. Partly in response to comments such as the ones made energetically by concerned reader NotGettingIt yesterday and partly in response to the author’s own fallible intuition, what I’ve done today is to change which sort of playoff odds are utilized in the team NERD scores. Until now, I’ve used the divisional-series odds from the FanGraphs Projections Mode. Starting now, however, I’ll use the divisional-series odds from the Season-to-Date Stats Mode.

The former, because it accounts for the actual players on each respective roster, is probably the better tool for estimating each club’s actual chances of qualifying for the postseason. The advantage of the latter mode, however — for this endeavor, at least — is that it reflects how an actual human person experiences the playoff race. Consider: the Giants currently trail the Dodgers by 2.5 games in the NL West. Because the latter club features a stronger roster than the former, their Projections Mode divisional-series odds are much better: 91.1% for the Dodgers versus just 25.6% for the Giants. What a human observes, though, is mostly just that 2.5-game lead — and a 2.5-game lead, generically speaking, doesn’t translate to such robust odds as those possessed by the Dodgers. The Season-to-Date Mode depicts more closely that generic translation: it gives the Dodgers a 70.5% chance of qualifying for the divisional series; the Giants, a 46.8% chance. In the former case, the two teams are separated by roughly 65 percentage points; in the latter, roughly just 25 points. While, again, the Dodgers probably possess something closer to 90% odds of qualifying for the divisional series, that figure is based on certain (sound) assumptions about the quality of their roster — assumptions with which a person, actually experiencing the event, is likely less concerned.

Below are now the top-five clubs by NERD. Note: pDOFF represents divisional-series playoff odds as calculted by means of the projections; sDOFF denotes the same thing, except as calculated by season-to-date stats. Team NERD formerly included the former; now, the latter.

# Club pDOFF sDOFF Old NERD New NERD
1 Giants 25.6% 46.8% 5 9
2 Pirates 49.3% 43.0% 8 9
3 Dodgers 91.1% 70.5% 5 9
4 Astros 49.3% 56.2% 10 7
5 Cubs 35.3% 23.9% 7 7

Given their greater proximity to the 50% DOFF mark by this new methodology, both the Dodgers and Giants’ NERD scores increase — which also has the effect of displacing the Astros to some degree, although Houston’s score remains high overall.

Readers’ Preferred Broadcast: New York NL Television.

One Other Brief Note
Today’s Free Game
Today’s free game features the same Dodger and Met clubs invoked above, and can be accessed by means of this hyperlinked text.

Complete Schedule
Here’s the complete and very sortable table for all of today’s games. Pitching probables and game times aggregated from MLB.com and also the rest of the internet. Note that the calculations both for team and game NERD scores have changed recently to better integrate playoff odds into same. Read more about those changes here, if you’re the sort of person accustomed to making poor life decisions.

NERD Image 4

Away   SP Tm. Gm. Tm. SP   Home Time
Ubaldo Jimenez BAL 6 5 6 5 8 NYA Masahiro Tanaka 13:05
His. Iwakuma SEA 6 4 5 4 7 DET David Price 13:08
Ervin Santana MIN 3 3 4 5 7 LAA Garrett Richards 15:35
R.A. Dickey TOR 7 6 6 5 6 OAK Scott Kazmir 15:35
Doug Fister WAS 3 4 6 9 9 PIT Francisco Liriano 19:05
Jeff Samardzija CHA 7 1 4 5 5 CLE Trevor Bauer 19:10
Clay. Kershaw LAN 10 9 7 4 6 NYN Bartolo Colon 19:10
Chris Young KC 2 5 4 5 4 STL John Lackey 19:15
Wade Miley BOS 4 4 6 7 8 HOU Lance McCullers 20:10
Mike Fiers MIL 3 5 5 6 5 AZ Zach Godley* 21:40
Tom Koehler MIA 3 6 5 4 7 SD Tyson Ross 22:10

* = Fewer than 10 IP, NERD at discretion of very handsome author.





Carson Cistulli has published a book of aphorisms called Spirited Ejaculations of a New Enthusiast.

18 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
woodge
8 years ago

I would have guessed that the new algorithm would favor a matchup between the two teams that are leading all of baseball in wins. Chris Young’s stats aren’t great but they’re almost good. And he is fun to watch, not because he is a dominating pitcher (though he has at times) but because he is such a statistical outlier–something that baseball Nerds love.

I suppose the true nerds are willing to analyze the NERD scores and pick their own games regardless of the algorithm.

I’m watching the I-70 series tonight.

someone
8 years ago
Reply to  woodge

Amusingly, that matchup actually has the worst game score of the day

tz
8 years ago
Reply to  someone
HappyFunBall
8 years ago
Reply to  woodge

You would have guessed wrong

Teams with very high or very low playoff odds are deemed to be less interesting than those in a tight race. Thus being a really really good team works against your NERD score.

woodge
8 years ago
Reply to  HappyFunBall

Ah, I see. Because no one wants to watch good baseball teams… /s

Well-Beered Englishman
8 years ago
Reply to  woodge

I, for one, am excited to watch the marquee matchup of Boy Wonder vs. Old Tub of Lard.

Wade Miley
8 years ago

Why are you calling me a tub of lard?