The most important of Miguel Sanó’s three true outcomes

Power has always been the name of the game with Miguel Sanó. Even as a 16-year-old, 190 pound teenage prospect (obviously before he was done growing), the draw was raw power. The scouting reports of him as an amateur shortstop glow of 80-grade power, optimism that he’d also hit for average, and acknowledgments that he’d quickly outgrow shortstop. Here’s an excerpt from the 2011 MLB.com report:

“He’s got a ton of raw power to all fields, and his outstanding bat speed should allow him to hit for average as well, though he’ll need to refine his approach at the plate. Signed as a shortstop, he’s split time between short and third as a professional and most feel he’ll end up being too big to stay up the middle”

Besides his bat speed and strength, reports from that period also pointed to the biggest question mark in Sanó’s profile — whether his aggressive approach would let him take full advantage of his power.

Here’s an excerpt from Baseball America’s 2011 report:

“The strength, bat speed, swing path and leverage are all there for him to hit 30 homers once he refines his approach and learns to recognize pitches. Like many young hitters, he sometimes struggles with spin…“

Naturally, with a powerful reputation that preceded him at every stop in the minors and reports suggesting he struggled with breaking pitches, pitchers carefully fed him a steady dose of spin in hopes that he would get himself out.

By 2013, when Sanó reached double-A, there was optimism and evidence he was developing in this area:

“Minnesota envisions him as a future cleanup hitter thanks to his present power, improved patience and pitch recognition. Sanó is learning to lay off breaking balls out of the strike zone…”

Despite his progress, by 2015 when he debuted in the majors, Sanó saw just 47% fastballs, the lowest rate of any hitter that took 300 plate appearances that season. He would navigate that better than anyone could have expected, posting a fantastic .269 / .385 / .530 line with an almost 16% walk rate in 80 games. In the offseason following that rookie year success, he was lauded by none other than FanGraphs as the rare slugger with elite plate discipline.

By now, you surely know, those proclamations might have been premature. Sanó has yet to match his rookie year production levels and has been anything but consistent season over season.

His time off the field has been anything but uneventful, as well, but that is a topic for a different post. Instead of becoming the rare hitter that hits for average and elite power, Sanó has become one of the poster children for Three True Outcomes — strikeouts, walks, and home runs — which we covered here last season.

Among the 359 qualified hitters since 2015, Sanó has had the third-highest percentage (54.52) of plate appearances end in one of those outcomes, trailing only Texas’ Joey Gallo (58.5%) and catcher Alex Avila (54.6%).

In hindsight, that outcome is not terribly surprisingly given the inherent uncertainty of projecting prospects and Sanó’s specific profile.

— — — — —

Often, when the three true outcomes label gets attached to a player, it is done derisively, alluding negatively to players with larger than life reputations from eras gone by like Rob Deer, Adam Dunn, or Major League’s fictional Pedro Cerrano.

But the rise in three true outcomes is a macro trend across the sport, reflective of the behavioral changes in how the game is played. The trend has accelerated in recent years thanks to the insights of Moneyball and advanced analytics improving our understanding of the relative run value of these events.

The proportion of all MLB plate appearances ending in one of the three true outcomes has risen steadily over the past 100 years. It first cracked 30% at the league level in 2012 and has risen even faster since, breaking 36% last season.

That trend has had negative consequences for the aesthetic appeal of the game, to be sure, but some of the best offensive players in baseball produce lots of three true outcomes. Aaron Judge ranks just behind Sanó in three true outcome rate, Bryce Harper is 21st, Mike Trout is 22nd, and Ronald Acuña Jr. is 26th. The top of the list also includes players more of the Deer and Dunn lineage, like Chris Carter and Chris Davis, but it is hardly an open and shut case that the three true outcomes archetype is a bad thing.

The primary difference between the Trouts and Acuñas and the Carters and Davises is that a larger fraction of the first group’s outcomes have been walks and homers, while the second group’s outcomes were strikeouts.

That brings me back to Sanó.

Let’s look season by season at how Sanó’s three true outcomes breakdown by type:

You can see the total percentage is trending upward. Look closely at the blue and green segments and you might see something interesting. In 2015 and 2019 slightly more than 20% of his plate appearances (add blue and green together) were home runs and walks. Those two years also happen to have been his most productive overall campaigns, with .392 and .378 wOBA, respectively. Those are numbers that are in the neighborhood of the Trout and Acuña camp.

In 2016, 2018, and 2020, his combined rate of home runs and walks was around 15% each season and his overall production was significantly lower (2016: .334 wOBA, 2018: .295, 2020: .317). Those are numbers that are in the neighborhood of the Carter and Davis camp. 2017 somewhat splits the difference with a home run plus walk rate around 17% and a decent .361 wOBA.

Below I’m overlaying his season by season wOBA (red line) with the percentage of his plate appearances that end in a walk or home run (black) to more clearly illustrate how related they seem to be. You can clearly see they track together:

Now, you might be thinking it’s not exactly groundbreaking to point out that he’s produced at a higher level when he’s homered and walked more frequently. Way to go, John. You solved baseball. You’re not wrong. But the degree to which his home runs and walks appear to drive his overall results is incredibly high. The correlation of the two lines above is 0.944.

In the two seasons when Sanó reached 20% of his PAs ending in a home run or a walk, he produced this composite line:

• .256 / .363 / .557, .384 wOBA, 35.9% K%, 14.0% BB%, 6.7% HR%

In the other four seasons, this:

• .233 / .314 / .466, .338 wOBA, 37.7% K%, 10.6% BB%, 5.3% HR%

The differences in the lines above are 23 points of batting average, 49 points of on base percentage, 91 points of slugging, and 46 points of wOBA. For comparison, among those 359 qualified hitters since 2015, .384 wOBA would rank as the 12th-best, while the .338 would be tied for 102nd.

Of the underlying stats, the walk rate jumped out to me most. Let’s investigate the free passes further.

— — — — —

Given that Sanó’s strikeout rate is easily the highest in franchise history among non-pitchers, you might have an impression that he has a free-swinging approach at the plate. It might be a logical assumption, but it isn’t true.

Since 2015, the average hitter has seen 3.9 pitches per plate appearance. Over the same period, Sanó has averaged 4.24. He’s been comfortably above average in that measure each season of his career.

Further evidence supporting the idea that Sanó has a deliberate approach at the plate is his 25.5% career chase rate, which compares favorably to the MLB average of 28.4%. While his propensity to chase has ticked up since his career best 23% rate in 2016, he’s been league average or better every season of his career. The chart below shows the distribution of single season out of zone swing percentages since 2015. Sanó’s seasons are shown by the red circles. The shortened 2020 season is excluded.

Not only does he not chase at a high rate, he simply swings less often than his peers. Nearly 47% of all major league pitches are swung at, yet Sanó has only swung at 44.1% of the pitches he’s seen in his career.

These points indicate to me that he takes a patient approach in search of pitches he can damage. He’s hunting so he can hit for power.

The challenge is, the pitchers know that, too. Given Sanó’s prospect pedigree, well known scouting report, and .256 isolated power (ISO, slugging percentage minus batting average) since 2015 (19th-highest), it’s a well understood fact that he can hurt pitches he can handle.

So pitchers just don’t throw him much to work with.

I mentioned above that he saw only 47% fastballs his rookie season. That wasn’t a fluke. It established the trend that’s continued through his career. Of the 9,572 pitches thrown his way, just 47.2% have been fastballs, a rate that places Sanó 355th out of 359 qualified hitters in seeing fastballs. Said differently, only four hitters have seen a lower percentage of fastballs in the last six seasons.

For good reason, too. Sanó has hit a combined .189 / .257 / .368 against breaking balls and a combined .175 / .261 / .385 against offspeed pitches (changeups, splitters, etc.). Both of those are significantly worse than the destruction he causes against fastballs — .292 / .392 / .610.

Beyond peppering him with breaking balls and changeups, pitchers are also careful not to throw Sanó many strikes. Since 2015, only 41% of the pitches thrown his way have been in the strike zone. Breaking that out by season, his rate was 44% in 2016 and has declined every season since, to just 38.1% last season. Comparatively, the league average last season was 41.2%.

The combined result of lots of breaking and offspeed pitches, mostly thrown out of the zone, is a pitch location map that looks like this:

Nate Silver, many years ago writing for Baseball Prospectus, found that there is a meaningful relationship between power and walks. Specifically, he found ISO to be a favorable predictor of non-intentional walk rate. I’d venture to guess that finding was due in large part to data similar to what we see above — pitchers carefully avoiding sluggers.

For Sanó, then, this means each plate appearance is a balancing act of aggressiveness and patience. He has to be patient to lay off the tempt offerings, aggressive enough to do loud damage when he gets something over the plate, and somehow be able to consistently discern the difference between the two. How well he straddles that fine line dictates how successful he will be.

Thanks to Statcast data and Savant, we can dig into how well Sanó has found that equilibrium.

— — — — —

The base-out run expectancy concept (RE24) has been extended to cover run expectancy by each of the twelve ball-strike counts (RE12). Combining the two, we now know the run expectancy by each ball-strike count in each base-out state (RE24xRE12, or RE288).

For example, the expected run value of a 0-0 count with the bases empty and no outs is 0.51 runs. Let’s say the first pitch of the plate appearance is taken for a strike, making the count 0-1 with no runners on and no outs. That situation has an expected run value of 0.47. Thus, we can calculate a difference of minus-0.04, which is debited against the batter. Let’s say the next pitch is put in play for a double. The expected run value of a runner on 2nd base, with no outs, and a 0-0 count is 1.15. The batter who hit the double would then be credited for 0.68 runs (the difference of 0.47 and 1.15) and his running total would be 0.64 runs (the sum of 0.68 and minus-0.04).

That data can be combined with pitch location data allowing us to analyze how batters perform against pitches in different locations in and around the strike zone. The Savant folks parsed the strike zone into four attack regions, shown below — heart, shadow, chase, and waste:

Given the map above, the definitions are pretty straight forward. About 25% of all pitches end up in the heart and batters swing at about 75% of them. If the batter doesn’t swing, the pitch in the heart will be called a strike. The shadow zone is targeted most often, with more than 40% of all pitches. The strike zone boundary splits the shadow zone, so, taken pitches in that area have about a 50/50 chance of being called strikes and batters swing at about 50% of them. Another 25% of all pitches are in the chase region. Pitches there are swung at about 25% of the time and will be called balls if taken. The remaining 10% or so are waste pitches that will be balls if taken and those are swung at just 5%.

This granularity gives us a useful framework for investigating Miguel Sanó’s balancing act. Conveniently, Savant has the Swing / Take tool that breaks down his run values by attack region. His summary data looks like this:

The chart above doesn’t show 2015, but if it did you would see that his total that year was +24, and he had positive run values in each of the four regions. I should also point out that these totals are a cumulative stat which means the 2020 figures are not adjusted to account for the shortened season.

A few things stand out. First, Sanó’s overall values (see “All” column) swing significantly season by season in the same pattern as his wOBA numbers did above. The total run values here correlate very strongly (r = 0.986) with his season by season wOBA.

Second, the shadow region has consistently given Sanó a lot of trouble, shown by the significant negative values every season.

Lastly, it’s pretty easy to see that his best seasons were driven by significant positive results in the heart, as well as the chase and waste regions — indications that he found a good balance of aggressiveness and patience. In contrast, his poorer seasons reveal much lower run values in the heart, chase, and waste regions — indications that he struggled to find the right balance in those years.

Each region can also be broken down by run value generated from swings and run value from pitches taken, which give us a way to validate those indications. In Sanó’s case, that data reveals something pretty interesting:

Overall, he’s generated far more run value from his takes than he has his swings. This makes sense since the majority of pitches he sees are balls, but the by season values here make it clear that he’s done a better job letting the bad pitches go by and going after the good ones in his higher production seasons. The data above follows the same pattern we’ve been seeing — 2015 and 2019 were his best years in terms of the balance between swinging run value and take run value and 2017 was pretty decent too.

How those swings and takes break down by attack region is informative, too. One last data table:

The only attack region where Sanó has a positive run value from swings is the heart (+86). To explain that in more familiar terms — Sanó has hit .376 and slugged .852 (!) for a wOBA of .503 against pitches in the heart. That includes a .327 average and .712 slugging percentage against breaking balls and .370 average and .907 slugging against offspeed pitches — significantly better production than what I showed you earlier against those pitch types.

In contrast, his swings in the shadow, chase, and waste regions have generated minus-148 total runs, with most of those (minus-88) coming in the shadow region. Again, to translate, he’s hit only .156 and slugged .271 against pitches in the shadow, waste, and chase regions. Those figures drop to just .118 and and .185 against breaking balls and .055 and .101 against off speed pitches in those regions.

— — — — —

To wrap this all up, I think the data tells a pretty clear story. Shrinking his attack zone and taking his walks is the key to success for Miguel Sanó. When he has done that in the past, he’s been a significantly better offensive player than when he hasn’t. Pitchers have shown they are content to throw him junk and lots of pitches in the shadow, chase, and waste regions — daring him to get himself out. He can’t hit those breaking balls and changeups out of the zone, anyways. So, he should let them go by and stay ready for them to bring him something in the heart. There, it doesn’t matter what type it is — fastball, breaking ball, or offspeed — he’ll hurt it. If they don’t come to him in the heart, he should happily take first base.

Of course, that’s easier said than done. But he’s done it before. He just needs to do it more consistently in the future. If he can, perhaps he still can make good on those glowing prospect projections from a decade ago.

---

John is a contributor to Twinkie Town with an emphasis on analytics. He is a lifelong Twins fan and former college pitcher. You can follow him on Twitter @JohnFoley_21