clock menu more-arrow no yes

Filed under:

Do scouting reports and stats projections match? A look at Royals prospects

New, comments

We're not selling jeans here.

Allan Henry-USA TODAY Sports

So we have to begin this article with the huge caveat that it's really, really, really hard to accurately project prospects, and especially those in the low-minors. There's a lot of noise in the signal. Most projection systems recognize that noise. From Baseball Prospectus:

In everyday boxscores, numbers represent outcomes: 2 BBs means a batter walked twice. But outcomes can be noisy: Maybe a bird flew in front of the pitcher just as he threw one of those ball fours, or maybe first base was open and the pitcher was on deck, or maybe the umpire was mad at the catcher, or maybe the batter is a 26-year-old in A-Ball and scares the wits out of the pitcher. What we want to do in sabermetrics is measure talent, or skills, based on those outcomes. Then, as best we can, we want to predict future outcomes based on the skills. Two walks in one game don't mean the batter has a good eye, and they don't mean he'll repeat the feat the next day. But if we can say the batter has plate discipline-and that the plate discipline was the underlying skill of those two walks-then we get more confident.

There's always going to be a disconnect between scouting and statistics. The classic scouts will live and die by what they have seen and what they can project by that. Sabermetrics looks much more coldly at the data and projects only what it can use statistically based on historical performances, aging curves, etc... One doesn't necessarily have to be better than the other.

For instance take Manny Machado. Scouts loved Manny Machado as did prospect websites. He was a unanimous top 10-15 prospect in baseball immediately after being drafted 3rd overall. Some scouts and websites were proclaiming him as a star already. That's a little hasty in my opinion, but the reports and scouting projections on him were highly regarded. Machado debuted in 2012 and was good, but only got about 1/4th a season under his belt. For the 2013 season, ZiPS projected him to be a basically average hitter (by wOBA) and put up a pretty good season (2.7 WAR) for a 20 year old. Machado ended up being worth more than 6 wins (based off an extremely good defensive year). This isn't a fault of the projection system, because 20-year olds are basically never that good and Machado only had 900 or so plate appearances in the minors to project from, but at the time it could be said that scouts saw him as an almost immediate star.

On the other hand, Paul Goldschmidt was never really liked by scouts. He was an all-bat 1B who was probably too old for his league, but ZiPS liked Goldschmidt as an above average hitter with a 30 HR projection in 2012. Joc  Pederson wasn't liked by scouts until he put up a 20/30 season in 2013 at AA. ZiPS projected him as a 2-win player prior to 2013.

Both scouts and projection systems liked Mike Olt. He was a top-50 prospect in in 2012 and 2013 (and subject to oft speculated trade rumors) and ZiPS liked him as near 3 win player for 2013. Steamer had an above league average wOBA. Olt's time was hampered by injuries and a concussion and has yet to bounce back to prospectdom. Scouts and projections like Jesus Montero too. The scouts and projections liked Brandon Wood, and Oliver projected him to hit 27 home runs in 2009, and Marcel had  .330 wOBA for him in 2008.

Basically projections/scouts either both get it right, one gets it wrong, or both get it wrong. This is the projection/scouting world we live in, and isn't that much more fun?

The point of this article is to see how the scouting reports match up to the projections.

Raul Mondesi

The general reports on Mondesi describe a hitter with potential for an average hit tool that's led by above average bat-to-ball skills and contact, but gets dampened by an aggressive approach. That's a thing we have to remember sometimes with Mondesi. His ability to make contact is very high, but he still has patience at the plate to learn, and unless you're Vlad Guerrero, even an elite bat-to-ball skill can be held down by swinging at pitches that don't need to be swung at.

Mondesi doesn't have wild strikeout numbers (highest K% coming in 50 games at Rookie level: 28%), but he's got three seasons with 22%+ K% (which would be above the MLB average). One important caveat of Mondesi is his age. When you're a 17 year old in A Ball, a 22% K-rate is excusable.

Some 2015 K% projections for Mondesi:

Proj Sys K%
ZiPS 26%
Steamer 25%

Those are above MLB average (bad in this case) K-rates. They aren't horrible, but Mondesi projects to be a low-OBP/Decent-average SS (think Alcides Escobar) to begin with and the strikeouts don't help his cause.

One thing that I really loved from that earlier referenced BP article was this:

That is the average decreased (weighted by PA's) in walk rate from minor league age/level to the majors. SO for instance on the extreme end, a 17 year old in foreign rookie averages ~4% decrease from his foreign rookie year walk rate to the major leagues.

Mondesi was an 18 year old in A+ in 2014, and we don't have a square for that one (there have only been a handful of 18 year olds in A+ to begin with the past few years), but looking at the squares around where his would be we could perhaps expect ~.60% decrease from his 5.1% walk rate (which isn't much at all in comparison).

Unfortunately we don't have the same chart for strikeout rates, but typically strikeouts don't go down as you go up levels (ala facing better pitchers in theory). Mondesi expects to stay right in the green groove of squares unless somehow he's still in A+/AA at age 21/22 (that would be 2-3 years from now).

Hunter Dozier

Dozier's initial reports was as a good-OBP/good-average hitter which was based on his strong pitch recognition skills, bat speed, and line drive swing.

From BP:

Leveraged swing; strength to produce hard contact to all fields; core is an asset; feel for barrel; sound strike-zone awareness; ability to put together quality at-bats;

His AA debut exposed some holes in his bat with inside pitches and upper velocity. That's something Dozier can actually work to correct and generally comes around with repetitions. Even with some of the struggles, Dozier still managed an 11.6% walk-rate, which would have put him in the top-10 or so walk rates in the Texas League if he qualified (267 PA).

If we go with the chart above, Dozier looks to lose around ~1% of his walk rate (12.35% in A+/AA). That would be an 11.35% walk rate (basically what he did in AA). That's still a top-25 walk-rate in the majors and about 50% better than league average.

Proj Sys BB% K%
ZiPS 7.9% 23.8%
Steamer 7.3% 24.5%
PECOTA 8.4% 25.7%

So there's a possible disconnect here. The projection systems only know about what Dozier has done (obviously) and what similar players at similar ages/leagues have done, then regresses a bit too. Dozier has certainly shown good walk rates in the minors, and walked in almost 1/5th of his plate appearances in the Arizona Fall League.

It's really a strong and maybe optimistic stance to project a prospect who hasn't played much in the high minors to walk at a top-20ish league rate, but Dozier has some track record of doing so and the historical averages above say it's possible too.

Bubba Starling

Where you waiting for this one? I kinda was too, so let's give us what we want.

Proj Sys BB% K%
ZiPS 6.2% 29.7%
Steamer 4.9% 29.3%
PECOTA 6.0% 30.8%

Pretty close numbers across the board. Below average walk rate, worse than average by a bunch strikeout rate. A 30% K-rate isn't dooming, but you've gotta hit basically 30 home runs and walk like 15% of the time... something Starling probably isn't going to do.

The numbers above also gel with the scouting reports. Starling is pretty well known for his hit tool, or lack thereof. He's one of those guys where a 40 or so hit tool would allow him to be an everyday player and possibly an above average one in the range of 3 wins or so given his power, speed, and defense.

I've talked about Starling in way too many digital and verbal volumes so I'm not going to rehash the same ideas and thoughts, but the projections and scouting reports are in line with each other.

The possible disconnect makes it easy for a player to change his projection. All he has to do is, well, perform well.. but which one is harder to do: perform well and change your projections or perform above your tools? The two aren't necessarily dependent of each other. One player can certainly outperform their tools, but not actually change their projection and a player can change their projection, but not their tools. Maybe it isn't which one is harder to do, but which one is more important?