:format(webp)/cdn.vox-cdn.com/uploads/chorus_image/image/60261413/927729472.jpg.0.jpg)
Back at the start of the season, I took a look at our Spring Training stats. I know that most ST stats are fairly meaningless but there was some research that suggested that certain inputs might still have some value. Specifically, BB% and K% can often help identify breakouts and busts.
For reference, here is the link.
So, let’s look at the data and see what we can find. This should be a fairly simple article.
Dbacks Spring Training Hitters K%
| Batter | ZiPS K% | K% | ST Trend | Current Trend |
|---|---|---|---|---|
| Batter | ZiPS K% | K% | ST Trend | Current Trend |
| Nick Ahmed | 19.40% | 20.60% | Worse | Worse |
| Alex Avila | 33.80% | 44.60% | Worse | Worse |
| Daniel Descalso | 22.00% | 25.10% | Better | Worse |
| Jarrod Dyson | 13.90% | 14.50% | Worse | Worse |
| Paul Goldschmidt | 22.40% | 26.30% | Worse | Worse |
| Jake Lamb | 26.30% | 24.30% | Better | Better |
| Deven Marrero | 28.80% | 27.10% | Worse | Better |
| Ketel Marte | 13.70% | 12.50% | Better | Better |
| Jeff Mathis | 31.10% | 30.00% | Better | Better |
| John Ryan Murphy | 18.80% | 27.60% | Better | Worse |
| Chris Owings | 21.80% | 26.00% | Better | Worse |
| David Peralta | 17.50% | 21.10% | Better | Worse |
| AJ Pollock | 14.70% | 22.90% | Better | Worse |
Hmm. There are a few ways to look at this. The first way is to count how many current trends matched the trend after Spring Training.
Number of matching current and Spring Training trends: 7
7 out of 13 hitters, or nearly 50%. As Jim mentioned in the previous article, we would expect to see this around 50% if it wasn’t indicative of any meaningful link.
However, the fact that 9 out of the 13 hitters are doing worse than their projections made me think: is there something these? We can spin a story for at least 8 of these hitters.
Look at Nick Ahmed, Daniel Descalso, John Ryan Murphy, David Peralta, and AJ Pollock. All of these hitters are enjoying career-bests in ISO and are 5 out of the top 6 for the highest ISO on the team. They could very well be selling out K% for power which is an overall benefit to the player (an increase in power is far more power than an equivalent rise in K%). You will notice that all of these players except Nick Ahmed did better than their projections in Spring Training.
And then there is Alex Avila, Paul Goldschmidt, and Chris Owings. These three are all substantially worse than their projections so far this season. Two of the three also performed worse in ST as well.
This might be an interesting coincidence but I think there is a plausible theory here: the players on the verge of breakouts might start seeing better batted ball results in Spring Training. The overall talent level is quite a bit lower in ST which makes it easier to make contact, thereby lowering ST K%.
I think this will be worth a look at the end of the season to see how things stabilize in a bigger sample. But I think there is a method here to be made for identifying changes next ST.
Now let’s look at BB%.
Dbacks Spring Training Hitters BB%
| Batter | ZiPS BB% | BB% | ST Trend | Current Trend |
|---|---|---|---|---|
| Batter | ZiPS BB% | BB% | ST Trend | Current Trend |
| Nick Ahmed | 5.70% | 7.40% | Better | Better |
| Alex Avila | 16.00% | 13.20% | Worse | Worse |
| Daniel Descalso | 11.10% | 13.50% | Worse | Better |
| Jarrod Dyson | 6.90% | 11.10% | Worse | Better |
| Paul Goldschmidt | 14.70% | 13.70% | Worse | Worse |
| Jake Lamb | 11.50% | 11.90% | Better | Even |
| Deven Marrero | 5.70% | 7.10% | Even | Better |
| Ketel Marte | 7.20% | 6.80% | Better | Even |
| Jeff Mathis | 6.00% | 13.00% | Better | Better |
| John Ryan Murphy | 6.80% | 5.90% | Better | Worse |
| Chris Owings | 4.90% | 7.80% | Better | Better |
| David Peralta | 7.40% | 8.30% | Better | Better |
| AJ Pollock | 7.30% | 7.10% | Better | Even |
Number of matching current and Spring Training trends: 6
Again, the first glance at the ST stats don’t really give us a good idea.
But let’s look at our two buckets of players. Let’s look at the breakout players (Ahmed, Descalso, JRM, Peralta, Pollock). 4 of them are currently exceeding or even with their projected BB%, as one would typically expect with a power breakout. 4 out of the 5 also exceeded their projections in ST.
Meanwhile, at the slow starters bucket (Avila, Goldy, Owings), 2 are currently doing worse now (again, as expected) and 2 out of the 3 were worse in ST.
There does seem to be a running theme. I’m going to break things down into buckets again, this time by their Spring Training stats:
Better ST K%/Better ST BB%: Jake Lamb, Ketel Marte, Jeff Mathis, John Ryan Murphy, Chris Owings, David Peralta, AJ Pollock
Three actual breakouts (JRM, Peralta, Pollock), one “hidden” exit velocity breakout (Marte), one bust (Owings), one looks-better-but-not-hitting-homers (Lamb), and one Mathis who doesn’t play much and still can’t hit.
Worse ST K%/Better ST BB%: Nick Ahmed
One more power breakout.
Better ST K%/Worse ST BB%: Daniel Descalso, AJ Pollock
Two more power breakouts.
Worse ST K%/Worse ST BB%: Alex Avila, Jarrod Dyson, Paul Goldschmidt
Three players that all badly underperformed and/or started the season slowly, with Goldy completely going nuts in June to (so far) salvage his season.
This is actually pretty interesting to me. I think there are things that can be accurately identified here out of Spring Training. I don’t think it’s going to be as simple as “If you have a better K% but worse BB% in Spring Training than your projections, you’re going to have a breakout” but I think there is meaningful separation between the breakouts and the busts. In particular, the busts really liked to make themselves obviously.
Keep in mind, this is just one team in one season of Spring Training. I need a lot more data to draw any statistical conclusions. But I will play this game again next season. And I think at an individual level, you can look into Spring Training K% and BB% (when combined with other data/factors) to maybe call out some breakouts/busts of your own.