By: Jake Kucheck
First and foremost, I have to get this out of the way… Jarrett Parker is back!
With two outs and the bases loaded in the bottom of the ninth, facing lights out closer Roberto Osuna, Parker strode to the plate down exactly four runs, with an opportunity grow his legend and make this particular writer’s curious Jarrett Parker obsession become ever so slightly more justified. And… he promptly struck out on three pitches, the last of which was roughly a 57 foot curveball.
Ok… so maybe Jarrett Parker won’t singlehandedly lead the Giants to a 2016 World Series winning the MVP and hitting 85 dingers in the process. In fact he’ll probably go right back to AAA once Angel Pagan is healthy. But Parker (and #EvenYears) aside, the Giants are off to a strange start. They seem to be either very good, or very bad, and lately, more of the very bad. But why?
Stats types (myself included) love to talk about variance, regression to the mean, and outliers. All of these things are mainly just fancy ways to say that unsustainable things are unsustainable, and that those things will eventually do what they did before they started being unsustainable. It’s the reason that Matt Cain threw a wonderful game tonight (he also stopped throwing the ball down the middle of the plate!), it’s the reason Buster Posey will not fail to get a hit for too many more plate appearances, and it’s the reason why Jarrett Parker could not continue his immaculate record of hitting a grand slam each time he strode to the plate with the bases loaded (yes, he only had one such previous plate appearance, but this statement is technically accurate).
Cluster luck, though, is more than just the ebb and flow of individual players. It measures a team’s performance in accruing (or allowing) consecutive hits, an event that is likely to produce runs. To give an example, let’s pretend a team is allowed a maximum output of 7 singles and 1 double at any point during a game. If you were a rational manager deciding when you’d like your singles and double, you’d allocate all of your singles consecutively and have your eighth hitter hit the double, meaning you’d score somewhere between 6 or 8 runs in the first inning. Conversely, if you were a terribly inefficient manager and scattered these 8 hits over 9 innings, you could very well get shut out absent some inordinate amount of walks and/or errors by the opponent.
Want a more real life example than the polar opposite hit allocators in that ridiculous hypothetical?
How about opening day? In that game, the Giants scored 12 runs, and everyone assumed that they would be a dynamic powerhouse of an offense combined with a star studded pitching trio that would steamroll its way to the 2016 World Series. What might have been more impressive, though, is that the Giants scored 12 runs on “only” 15 hits. How did they do that? Cluster luck. Here are each of the innings paired with the number of hits in that inning:
1-0, 2-2, 3-3, 4-1, 5-2, 6-0, 7-0, 8-6, 9-1.
What should be obvious is that the Giants scored the most runs in the 8th inning of that game, but also, that they had 2 innings of 3+ hits (an inning very likely to result in a run or runs). That kind of grouping of hits, especially when combined with plate discipline (resulting in walks/extra baserunners) and power (ability to optimize impact of hits), will be a recipe for success more often than not.
By contrast, we should also examine what happens when cluster luck is not working in a team’s favor. A recent example leaps off the page, with the Cinco de Mayo slugfest that ended with the
Denver Broncos Colorado Rockies beating the San Francisco 49ers Giants 17-7. In this game, the Giants pounded out one more hit than their opening day barrage, but managed “only” the 7 runs. Here’s their inning/hit line from that game:
1-3, 2-2, 3-2, 4-1, 5-4, 6-1, 7-0, 8-2, 9-0.
That doesn’t look too bad. But we should also consider what the Rockies were able to do on offense, with only one more hit.
1-4, 2-2, 3-0, 4-1, 5-10, 6-0, 7-0, 8-0, 9-0.
In terms of efficiency, the Rockies 5th inning set a franchise record for number of runs in an inning, putting up 13. Put another way, the Rockies were nearly twice as productive in the 5th inning of that game than the make believe team in the hypothetical scenario I proposed that you were considering ridiculous only a few paragraphs ago.
What we know is that the Giants are not as good as they were on Opening Day, and the Rockies are not as good as they were in the fifth inning of last Thursday’s game. We know this because of our friends variance, regression to the mean, and outliers. Cinco de Mayo was certainly an outlier, although 3 of the 4 Rockies wins in which they have put up 10+ runs have come against the Giants (all Peavy/Cain starts, unsurprisingly).
So just how unlucky are the Giants on the whole this season? According to ThePowerRank.com, they are the 23rd luckiest (or 7th unluckiest) team overall, combining both cluster luck and cluster luck allowed. Since no one thinks the Giants are the 23rd worst team in baseball, odds are that we will see their ability to string together hits (and scatter hits on defense) improve. We might even see a 15 hit, 12 run performance like the one on opening day.
Things will get better, there’s no question, but will it be too late? The silver lining here is that the Dodgers, Diamondbacks, and Rockies have also been unlucky so far, ranking 21st, 26th, and 28th respectively in cluster luck. Interestingly enough, the Padres, who everyone assumes are terrible, have parlayed their 4th luckiest rank to a spot while still in last place in the NL West, only 4 games behind the division leading Dodgers.
What does all of that mean? Mainly that there is a lot of baseball left to play, and right now the NL West is completely up for grabs. Should be fun to watch.