Fun With Projections: Stolen Bases

Dan Richards identifies players you should target and avoid for speed.

Projections aren’t the end all be all. No, Mike Trout likely won’t hit exactly .299 with 41 R, 16 HR, 37 RBI, and six steals over 239 PA (pursuant to ATC projections). Still, projections are ultimately predictions—the system’s best guess at a player’s highest probability outcome when, in reality, players are subject to a wide spectrum of outcomes of varying degrees of probability.

Projections are useful for a variety of reasons, chief among them crafting personal rankings. At a glance, however, they also illustrate mismatches in players’ draft prices. Said another way, if a projection system tells you OF50 and OF75 are likely to have highly similar outcomes, then you should probably just wait and draft OF75.

Let’s perform such an exercise, but for speed targets only. Say you’re uncomfortable spending substantial draft capital on Trea TurnerAdalberto Mondesi, or Jonathan Villar—as many drafters are. Who should you target if you want to stay competitive in the SB category? Who should you avoid?

Projections hold the answer.


Middle Rounds


I won’t exhaustively detail my thoughts on every player projected to steal some bases. Instead, let’s say you’ve missed out on the big-name, early-round base stealers. Now, you’d like to target more balanced hitters who will keep you competitive in the category. Maybe you’ve got your eye on some available shortstops like Bo Bichette and Tim Anderson.

Bo Bichette 49.1 211 31 8 25 7 .275 1.57
Tim Anderson 99.8 216 30 7 25 8 .279 2.47

Before comparing the two, some preliminary observations are in order. First, the ADP data is from 15 NFBC drafts since the announcement of the 60-game season. Second, these projections are from ATC.

Third, the Adjusted AVG (AAVG) column lets you see the true impact of a hitter’s batting average. A hitter’s AVG—a rate statistic—will be more or less valuable depending on how much it impacts your team. The more at-bats a hitter has, the more it impacts your team. That can be a good thing, as in the case of Jeff McNeil who projects for 203 AB and a .294 AVG. Or it can be a bad thing, as in the case of Joey Gallo who projects for 175 AB and a .232 AVG.

To be sure, AAVG depends on the league average batting average. By that, I mean the average of your league’s hitters’ batting averages, not all hitters in MLB. A player’s AVG will only hurt you in the category if it’s lower than your league’s average, and vice versa. For the purposes of this exercise, then, I performed my own mock draft, found a league batting average, and then determined Bichette’s and Anderson’s AAVG based on how their projected batting averages stacked up to the league average, multiplied by their at-bats.

With that out of the way, let’s return to the above table. Bichette projects for one more Run and HR, while Anderson projects for one more SB and an AVG that is four points better (over five more AB). Both guys should leadoff for respectable lineups per Roster Resource (although I would argue the White Sox lineup is more potent than Toronto’s), they play the same position, and their ages are already factored into their projections. Notably, Bichette hit the ball much harder than Anderson last year, posting a 6.1 Brls/PA%, 92.9 mph exit velocity on flies and liners, and 43.5 hard-hit%. Compare that to Anderson’s measly 3.9 Brls/PA%, 90.7 exit velocity on flies and liners, and 37.3 hard-hit%.

Nevertheless, Anderson managed a .294 expected batting average last year, compared to Bichette’s .273. Anderson also stole 17 bases in 518 PA (3.3 SB/PA%), whereas Bichette only stole four in 212 (1.8%). Of course, Bichette has a track record of running in the Minor Leagues, but there’s no guarantee that will continue this year, whereas Anderson has shown a proclivity to run since his call-up. It’s no surprise, either, given Anderson has the better sprint speed (28.7 ft/sec) even though he’s five years older than Bichette (28.4 ft/sec).

Ultimately, Anderson and Bichette have nearly identical projections, but one gets drafted, on average, exactly 50 picks later. You can quibble with the projections by increasing Bichette’s HR/RBI (or reducing Anderson’s), but to be intellectually honest, you should then also increase Anderson’s AVG or SB (or reduce Bichette’s). There’s a lot to like about both players, but if you’re looking for speed, take advantage of Anderson’s discount.


Late Rounds


As the draft progresses and you’re growing more desperate for steals, don’t fall into the trap of just sorting the remaining players by projected stolen bases and grabbing the top guy according to ADP. Strike when the time is right.

No pair of outfielders better represents that pitfall than Oscar Mercado and Lorenzo Cain.

Oscar Mercado 115.9 197 28 6 21 8 .259 -1.69
Lorenzo Cain 184.5 203 28 5 19 7 .279 2.32

Mercado projects for one more HR, two additional RBI, and an extra stolen base. However, Cain has a twenty-point-higher projected average spread over six more at-bats. That explains why his AAVG is in the black, whereas Mercado’s is in the red. Next, observe that, in terms of raw power, these are two extremely similar hitters.

Name Avg Launch Angle EV on FB/LD Hard-Hit Rate Brls/PA
Oscar Mercado 13.4° 90.9 mph 33.9% 3.1%
Lorenzo Cain 6.6° 91.6 mph 42.1% 3.0%

Mercado puts the ball in the air more, but Cain hits it harder. That’s how Cain is able to maintain almost the same barrel rate as Mercado despite such a low average launch angle. And the low launch angle is how Cain sustains a higher batting average.

Indeed, ground balls and line drives run higher BABIPs than fly balls and pop-ups, and Cain’s 50.9 GB% and 27.2 LD% in 2019 were both well above league average according to Statcast. Likewise, his 16.7 FB%, 5.2 PU%, and 27.4 Pull% were all below league average. Such a batted ball profile explains the high batting average projection, as well as Cain’s .290 xBA last year.

Conversely, Mercado’s mediocre .262 xBA is no surprise in light of his below-average 41.9 GB% and 24.9 LD% and above average 25.2 FB%, 7.9 PU%, and 40.0 Pull%. Basically, Mercado is trying to pull fly balls to maximize his power, but that also incidentally minimizes his expected batting average. That’s not what I look for in this type of player.

Admittedly, Mercado’s 29.5 ft/sec sprint speed dwarfs Cain’s 27.8 ft/sec mark, and likely explains why Mercado’s xBA didn’t drop even further than .262. However, as with Anderson and Bichette, what matters more is whether runners take advantage of their stolen base opportunities. Cain stole 18 bases in 623 PA last year (2.9 SB/PA%), whereas Mercado stole 15 in 482 (3.1%). All that additional speed didn’t translate into substantially more value for Mercado owners.

As before, these are two hitters who play the same position and are both projected to hit second for their teams. Fundamentally, they project similarly, with at least arguably a slight edge for Cain based on a stronger batting average. So why not grab a power bat, closer, or starting pitcher around pick 116, and just take advantage of Cain’s discount?

Photos by Dan Sanger/Icon Sportswire and Keith Allison/KeithAllisonPhoto.com | Adapted by Rick Orengo (@OneFiddyOne on Twitter and Instagram)

Dan Richards

Dan is a lifelong New York Yankees and Giants fan. A practicing attorney, Dan is better known for aggressively bothering his leaguemates about trades. You can follow him on Twitter @Fantasy_Esquire or by clicking the button above.

Leave a Reply

Your email address will not be published. Required fields are marked *

Account / Login