It's been four months since I published the first entry in the Scouting Grades series, and it's fair to say that motivation to continue blogging about baseball has been limited in that time. With the 2020 draft in the rearview, I found myself thinking about scouting grades again.

If you haven't already, it's probably a good time to review the post on Speed. Now, let's talk about POWER.

Power wins games. Power puts butts in seats. No other tool has a showcase -- the home run derby -- that exists specifically to elevate and celebrate a specific tool. It should come as no surprise that power also draws big signing bonuses and buys prospects more time to develop.

In this context, development most often means translating raw power into game power, the actual production of runs. That development generally takes the form of increased plate discipline which is not limited to swinging at fewer balls but also includes learning and taking advantage of pitch locations that really allow the hitter to get to that power.

Raw Power vs Game Power

Before player- and ball-tracking technology, raw power grades were determined by watching a hitter unload on baseballs -- usually during batting practice, a scout's best opportunity to see a lot of swings in a short amount of time. Raw power was essentially how hard a player could hit the ball, frequently evidenced by distance.

[Note: batting practice is not the only input, obviously. Game swings matter more and commonly result in harder contact than batting practice swings. However, scouts will almost always see fewer game swings than BP swings, and plenty of hitters will show you what they've got during BP.]

The Raw Power Scouting Scale (roughly):

• 20 – Even with a favorable gale force wind, the ball probably isn't going over the fence.
• 30 – A stiff breeze should do it.
• 40 – Doesn't need help, but you don't know it's over the fence until it is.
• 50 – Hey, there's some pop!
• 60 – Stadium vendors and people on the concourse should pay attention.
• 70 – The ball hits weird parts of the stadium, and occasionally leaves to check out other places.
• 80 – The ball has a greater than zero chance of escaping Earth's atmosphere.

Game power is an assessment of how often a player is going to tap into that raw power against MLB pitching. There's no escaping the subjectivity required for a game power grade since it relies on, among other things, the player's future hit tool and future raw power which are both projected by the scout.

This produces an unbelievably wide spectrum of projections and results. There are guys you've never even heard of with top shelf, 80-grade raw power that projected to hit approximately 0.000 at the highest level. Then you've got freaks like Frank Thomas and Miguel Cabrera that are just as likely to win a big league batting title as they are to win a homerun crown.

The Game Power Scouting Scale (roughly, thrown into disarray by the new baseball; not funny):

• 20 – 5 homeruns in a single season would be incredible.
• 30 – ~5 homeruns per season.
• 40 – ~11 homeruns per season.
• 50 – ~18 homeruns per season.
• 60 – ~26 homeruns per season.
• 70 – ~35 homeruns per season.
• 80 – Homerun crown contender.

Of the obvious variables that limit raw power in games -- pitch recognition, plate discipline, swing length, and others which will be discussed in the Hitting post -- there is one in particular that can limit a hitter with plus hitting and plus raw power to mediocre or even below average production: launch angle.

Exit Velocity & Launch Angle

Three names immediately jump to mind when I think about how a low launch angle stifles power: Nick Markakis, Eric Hosmer, and Yandy Diaz. That's three big dudes with big raw power, decent batting averages, and just not very many homeruns. Depending on your age, you may recall that, as prospects, Markakis and Hosmer were fairly widely believed to be potential perennial 40-homerun hitters. Diaz, who arrived just in time for launch angle to become a hot button issue, has been frequently noted as a guy that hits the ball as hard as anyone else in the game.

Last season, all three were in the Top 50 in Hard Hit % (95+ MPH) -- Hosmer #32, Diaz #43, Markakis #48 -- just ahead of Mike Trout (#49) and Fernando Tatis, Jr (#50).

Last Season, all three were in the Top 60 in Average Exit Velocity -- Diaz #20, Markakis #30, Hosmer #57 -- ahead of Anthony Rendon (#64). Trout checked in at #51. (Stats according to StatCast via Baseball Savant.)

In 469 plate appearances, Markakis hit 9 homeruns. In 667 plate appearances, Hosmer hit 22 homeruns. In 347 plate appearances, Diaz hit 14 homeruns. That's 45 homeruns in just under 1,500 plate appearances.

In 600 plate appearances, Trout hit... 45 homeruns.

Given 2.5x the number of plate appearances, three hitters that are ostensibly better at hitting the ball hard than Mike Trout combined to produce the exact same number of homeruns as Mike Trout alone.

The key difference (and admittedly not the only difference) is launch angle.

Eric Homser had a 2.1° average launch angle. Yandy Diaz had a 5.7° average launch angle. Nick Markakis had a 7.3° average launch angle.

The average launch angle for all MLB batted balls was 11.2°.

Mike Trout had a 22.2° average launch angle.

The point of all of this is to illustrate that translating raw power into game power requires hitting the ball in the air. A player can hit 100 MPH ground balls all day, but it won't result in power production.

Context and Projection

As scouting evolves thanks to the ever expanding use of technology, the scout's role will increasingly be to provide context for objective data rather than authoring almost entirely subjective reports. With the proliferation of Trackman, HitTrax, and similar systems, for example, it's a matter of course to have objective data for a player's exit velocity and launch angle. A scout provides little of use by assigning a 20-80 grade for present raw power.

A player's future raw power, on the other hand, is a contextual grade usually backed by an assessment of physical projection. Physical projection offers insight into a player's potential through continued growth and added speed/strength. An 18-year-old that could still grow a couple of inches and has never lifted a day in his life could get a 2-grade bump, while a 23-year-old with a mature build generally gets none.

When it comes to game power, the hit tool is the big separator. It is easily the most complex tool grade and is frequently broken into several sub-grades, and the realization of hitting potential often relies on additional external factors that are not easy for a scout to assess. That really is a topic for a future post, but a few quick examples will illustrate the type of context scouts should be chasing.

The path of the barrel, including its overall length and depth, plays a key role for both power and hitting. While a long, flat path might lead to a lot of overall contact, it isn't likely to lead to much hard contact. A shorter path on plane with the pitch has a better chance of producing hard contact, and being on plane with the pitch has the added benefit of an increased launch angle.

The length of the swing ties into timing which has downstream effects on pitch recognition and plate discipline. All other things equal, a hitter with a longer path to contact has to start to swing earlier, giving the hitter less time to see the ball before launching a swing. Less time to see the ball means worse recognition which will lead to bad swings and bad takes.

When timing is more easily disrupted, the contact point "moves" to different parts of the swing. Generally speaking, if the swing is late, the ball is more likely to be popped up because the point of contact has "moved" deeper where the barrel hasn't come back up to the anticipated point of contact yet. If the swing is early -- particularly on slower pitches that drop more -- the ball is more likely to be hit on the ground because the point of contact has "moved" out front where the barrel has already risen past the anticipated point of contact.

These issues are amplified more in swings that are further from the plane of the pitch.

Projecting a future game power grade can seem a lot like witchcraft, but it boils down to just a couple of questions:

• Will the player's raw power increase or decrease?
• How will the player hit against top-level pitching?
• Will the player hit the ball in the air?

At this point, projecting a player falls into the gap between scouting and player development. Amateur scouts can get to know a guy well enough to have a good idea of the player's development potential ahead of the draft, but the same can't be always be said for pro scouts who are watching another ball club's players. This development gap is a complex topic and will hopefully be the focus of a future post.

Some grades are straight-forward, objective evaluations, but others are quite subjective and open to varying degrees of bias. The Scouting Grades series aims to discuss different tools and how they are evaluated.

Speed is one of the more objective evaluations, but it isn't as straight-forward as one might think. Let's start with a fairly common chart that some of you have probably seen before. This chart contains the generally accepted guidelines for objectively converting a batter's Time to 1B into a scouting grade for speed.

Grades are typically meant to represent a normal distribution centered around 50 as MLB average (or 5 on a 2-8 scale) where 40 is 1 standard deviation below average and 60 is 1 standard deviation above average. A quick look at the StatCast baserunning sprint speed numbers for 2019 blows that idea out of the water.

128 batters averaged 4.55+ seconds on competitive runs to first base. That's almost exactly 25% of the 510-player sample objectively sitting at or well below the bottom of the scale. The median time of 4.38 seconds is roughly a 35 according to this scale. For comparison, normally distributed speed grades would put 68.2% of players between 40 and 60, with an additional 15.9% above 60, leaving 15.9% below 40, and only 2.1% below 30!

I took a quick crack at dividing the Savant list into separate tabs for left, right, and switch hitters and posted it in a Google Sheets document - 2019 MLB Baserunning Speed. I may have missed a leftie or two (#manualData) as right-handed (the default starting point), but a few lefties sneaking into the righties data won't ruin the samples. Here's what the chart would look like based on the split data with 333 right-handed batters and 132 left-handed batters. (44 switch hitters were left out since their times were presumably mixed.)

This chart is a more accurate representation of the actual Time to 1B for MLB players with the one exception being that no one would qualify as an 80 runner.

MLB clubs have undoubtedly been aware of this disconnect since stopwatch times were first compiled on a spreadsheet. An educated assumption here would be that clubs are more or less ignoring current speed grades in favor of objective measures from the player tracking technology deployed across ballparks all over the baseball world. Yet the guidelines remain in effect in most scouting contexts.

Where the scouting element may actually come into play is in projecting future speed. More on that later.

The Before Time / The Long, Long Ago

Before player tracking systems were everywhere, measuring every runner on every play, the only way to get objective speed measurements came from scouts' stopwatches. Time to 1B is a standard because it is a fixed-distance sprint that is run by every batter.

To properly capture a Time to 1B, a scout anticipates contact, attemping to start the stopwatch at the exact moment that bat hits ball, and then reads the batter's steps, attempting to stop the stopwatch at the exact moment the batter's foot touches the base.

As error prone as this might sound, you may be surprised to learn that scouts actually get pretty good at this with practice. Is it as good as a player tracking system? Obviously not. Does it get the job done anyway? Somewhat surprisingly, yes.

Speed elements and application, the downside of relying on Time to 1B

What does Time to 1B really tell you about a player's speed? It doesn't really tell you his peak speed, and it doesn't really tell you how quickly he accelerates either. What it does tell you is a decent approximation of the two.

Take another look at the spreadsheet I prepared. You can get a decent idea of who accelerates well by looking for players that are faster to 1B than their sprint speed peers. For instance, it's pretty interesting to see Jeff McNeil average a 4.10 with average sprint speed.

For Time to 1B, peak speed would need to be significantly faster to make up for below average acceleration, given the relatively short sprint. Peak speed really comes into play over longer sprints: gap fly balls, doubles and other two-base sprints, and especially triples. In other words, Player A might be slower than Player B to 1B but faster than Player B to 2B on a double. Does Time to 1B actually tell you which player is faster?

Speed impacts defense almost purely as it relates to range. Outfield defense typically falls into the same acceleration-and-peak mix as baserunning, but for most infield defense, acceleration is far more important, with peak speed generally only coming into play on pop-ups in No Man's Land.

But there's something else about Time to 1B that you may have not considered yet. Different batters have different swings and require different adjustments to transition from swinging to sprinting. Balance, momentum, stance, and overall effort each affect that batter's ability to recover from the swing and get moving toward 1B, and every batter's Time to 1B has this swing effect rolled into it. 2 players could have the exact same acceleration and peak speed but different Times to 1B because of different swings!

There is no swing effect on defense, and that may be the only thing that really prevents Time to 1B from acting as a near-perfect proxy for an outfielder's defensive range. For infield range, Time to 1B would seem to have little or no correlation. (NOTE: I'd enjoy looking at a study that digs into this idea, and I'll cover range more completely in the Defense entry in this series.)

Projecting Future Speed - A Case Study

This could probably be an entire series of articles by itself, so we're going to floor it for a bit, then slam on the brakes, get out of the car, and do a walk-around.

The one thing to keep in mind is that, absent an alternative guideline, a speed projection should target the player's speed at physical maturity, not the end of the player's career. Physical projection plays an immensely important role in projecting speed.

Nomar Mazara made his Double-A debut with Frisco late in the 2014 season. He was 19 years old and had a wire frame on which you could hang a lot of mass. He showed coordination but lacked any sense of athletic explosiveness. He routinely ran 4.70+ to 1B.

He was reportedly 6' 5" when he signed (source) as a 16-year-old and was now listed at 6' 4", so a scout could fairly assume that Mazara had been that tall for at least 3 years, all while working with professional strength coaches and trainers. By all accounts, he would get stronger and heavier as he got older.

Mazara was a 20 runner with a profile that screamed for a negative speed projection, so of course he returned to Double-A Frisco in 2015 running sub-4.40.

There's a brightside, though. Having jumped two grades in one off-season, there was now room for negative projection! That may sound like a joke, but sticking with the negative projection on the 2015 report is the right call. The better Time to 1B obviously indicates more explosiveness, but everything else in the projection is still true.

If you thought he was going to be a large, lumbering fellow at maturity in your original projection, the only difference in the new projection would be some lighter lumbering.

What else could this two-grade jump indicate? It takes a lot of work to jump a grade in anything, and Mazara jumped two grades in a single off-season! A scout would be crazy to positively project Mazara again, but the jump alone is arguably enough of an indicator that the negative projection should be smaller than originally projected.

Positive speed projections are extremely rare outside of young, undeveloped athletes. In Mazara's case, it seems reasonable to conclude retrospectively that Mazara still fell into that categry, but between 2014 being his third full year as a professional and the rarity of positive projections even within that category, a positive projection would have been met with skepticism.

TL;DR

• Scouting for speed is probably going the way of the dodo -- if it hasn't already -- thanks to tracking technology that puts a scout's stopwatch to shame, but physical projection is still the scout's domain.
• A competitive Time to 1B -- measured from bat-on-ball to foot-on-bag -- represents a good-enough estimation of most practical applications of speed in baseball.
• Projecting speed is the art of projecting physical maturity against present ability.