Using Change% to Quantify Batter Clutchness

Written by: Carlos Marcano (@camarcano)

Follow us on Twitter! @Prospects365

How many times have we heard or thought during a game that X player is a clutch batter? Or, on the contrary, that he tends to choke when needed the most? These comments are most of the time based on narratives we, as observers, create from specific situations that get fixed in our memories and bias our judgment; as our brain needs to organize the information we get through our senses in patterns, we easily label single events as part of a bigger sequence, whether that is true or not, and that’s why, for example, if we saw said batter hitting a walk-off single or got stricken out to end a rally recently, we immediately generalize these high impact situations as trends, when they are not, and these trends fog our perspective.

Clutch is a hell of a hard thing to objectively describe or measure and even harder to predict, if possible at all, not that huge and great efforts have not been made to address that. Clutch, LI, WPA/LI, and other stats are part of the tool set that brilliant analysts have devised for it, and they usually do a good job. But, sometimes, those stats might feel hard to grasp, not because they are too abstract or hard to understand but because they typically involve a lot of smaller calculations to obtain a result.

In my perennial journey for simplification, I wanted to try a simpler approach to this concept and I decided to focus on one question: What is the ultimate goal for a baseball team? There could be very long answers to this but the short and simple one is “to win”. That’s it.

So, when does a team achieve this?

A team can’t lose when it is tied or ahead of the opponent. Then, one of the most important things a batter can do for his team is, effectively, to tie or to put his team ahead whenever the chance presents itself, in other words, whenever he can change the possible outcome of the game for his team’s benefit. These players are, on those occasions, true Game Changers.

Under this premise, I decided to look for those batters that, with any part of their offense game, produced RBIs to tie the game or to put his team ahead this season. I went to Stathead, pulled the data for every player that meet those criteria, and summarize it. In total, 424 batters on any instances of their teams’ games were Game Changers and this is the list, sorted by the most changes:

Not the guy we were expecting at the top, right? With a .240/.355/.421 line, a .335 wOBA and wRC+ of 115, well, that’s just not a profile of a batter we would want to take the at-bat for our team when we need the necessary turn of events that would change the score on our favor. But in fact, that’s what he has done more than any other player so far.

José Abreu, Mike Yastrzemski, Brandon Lowe, and Starling Marte share the second place with 14 game changes each; the first three of them are having a great season offensively while Marte is more in line with Seager but still better overall.

Names like Mike Trout, Fernando Tatis Jr., Mookie Betts, Freddie Freeman, Manny Machado and/or Francisco Lindor are lower in the chart that what the consensus might guess. Could it be that as we are looking at the absolute values we are not taking into account that some batters might have more opportunities to tie or put their team ahead than others, so we are getting a skewed result?

I had to swim in the data and pull more than 23,600 Plate Appearance records where the batters had the opportunity to tie or put his team ahead. After sorting and refining, this is the list I got:

Batters with more than 35 chances made the cut for the list (average this season is approximately 68 for this data set).

There are a few things that are pretty interesting and probably surprising about these figures:

  • Batters have only been able to change the outcome under these conditions around 1 in 5 chances they’ve had, at best, and 8% of the time on average. That means that this type of game changes are rarer than one could expect.
  • We have to add “opportunism” as another attribute in the resurgence of Byron Buxton, as he is the batter that has taken more advantage of the opportunities he’s had to change the game.
  • This is another aspect in which Kris Bryant’s season is truly hideous: in 57 chances he hasn’t been able to make a single change. Not once. Zero.
  • José Altuve, JD Martínez, Pete Alonso, Eugenio Suárez, Anthony Rizzo, Anthony Rendón and Paul Goldschmidt are all big name players that are under performing here as their Change% is below the league average.
  • Kyle Seager, Dominic Smith, José Abreu, Mike Yastrzemski, and Cory Seager are still as good, relatively to the rest of the players, when measured by their Change% as by the total Changes they’ve done.

These snippets are fun facts and interesting ideas that will for sure add value to the discussions when analyzing this season, as a deeper dive in the data will always reveal things that we might take for granted, or we might not be acknowledging. But today I want to challenge it in a more empirical manner: I want to try to use this information to predict some type of value.

I am in no way saying that this data has prediction capabilities, don’t get me wrong. There are too many variables that I am not taking into account and I’m not controlling for so it would be disingenuous to say otherwise. I just want to make an exercise with what is available and find out in a practical manner if there is a bigger opportunity beyond what it looks like.

So, as we are approaching the Playoffs, I would like to boldly predict which player(s) will earn any of the MVP awards for the Playoffs. The spots for that stage are not completely filled yet but at the moment of writing this, the AL participants are:

Standings updated after games on Sept. 17:

  1. Rays (E1), 36-19, .655
  2. White Sox (C1), 34-20, .630
  3. A’s (W1), 33-20, .623
  4. Twins (C2), 33-22, .600
  5. Yankees (E2), 31-23, .574
  6. Astros (W2), 27-27, .500
  7. Indians (WC1), 30-24, .556
  8. Blue Jays (WC2), 28-26, .519

And for the NL:

  1. Dodgers (W1), 38-16, .704
  2. Cubs (C1), 32-22, .593
  3. Braves (E1), 32-22, .593
  4. Padres (W2), 34-20, .630
  5. Marlins (E2), 28-26, .519
  6. Cardinals (C2), 26-25, .510
  7. Reds (WC1), 28-27, .509
  8. Phillies (WC2), 27-27, .500

I will pick a player from most of these teams (and some ‘outside-looking-in’ contenders), according to their Change%, and those will be the candidates to win the awards.

The nominees.

These charts, one per league, summarize the candidates:

Logically, Buxton and Moreland are the top candidates to win it all, in terms of playoffs awards.

Why do I think that Change% could potentially shed some light on this topic?

It’s all about narratives.

If Change% let us discover batters that maximize, in some way or other, their approach to chances then said batters will “appear” in the moments that fans and analysts tend to remember the most: those special circumstances when things turned around for better for the player’s team because of him, and that narrative will be important when people cast their vote for the awards; this is especially important for short term scenarios like the playoffs are.

That’s what we are betting on here.

Follow P365 MLB Analyst Carlos Marcano on Twitter! @camarcano

Follow us on Twitter! @Prospects365

All statistics from Stathead, Baseball Savant and Fangraphs

Featured image courtesy of FOX Sports


Leave a Reply