clock menu more-arrow no yes

Filed under:

The Pitchers

New, comments

The following describes what I do to edit the stats of the pitchers in MLB: The Show.

Harnessing his control will turn Alex Meyer into one of the best pitchers in the Twins organization.
Harnessing his control will turn Alex Meyer into one of the best pitchers in the Twins organization.
Reinhold Matay-USA TODAY Sports

This whole train wreck of an idea of "fun" started way back with Carlos Silva when I noticed he was lacking his signature sinking fastball. As I've gotten older, I've started to notice that pitching fascinates me more than anything else in the game. It's a shame that the Twins have lacked some exciting pitchers over the past few years and I wish I could have appreciated Johan Santana and peak Francisco Liriano more, but hopefully the new crop of pitchers the Twins call up shortly will satiate my desires.

But we're here to discuss video games. What I've done is find - or at the very least attempt - a way to accurately represent each of the stats a pitcher has in The Show. I'll admit that in some places I cut corners, but I'll give a reason for each when applicable.

Before I get to the stats themselves, I need to talk about creating a purely average player. That's right, I went in and made a player with 50s in every possible stat. I even gave him the perfectly average name of Joe Smith (not to be confused with actually-good Angels reliever Joe Smith). Perhaps school has corrupted me into thinking that being in the 70% range was average, but according to The Show, this is actually what you get with a purely average Joe Smith as a starting pitcher, reliever, and closer.

Overall Rating of Joe Smith, The Perfectly Average Pitcher

SP - 63, RP - 60, CP - 60

Rather disappointing, I must admit. Well, step two was to investigate how each stat impacted a pitcher's overall rating. Yes, I really did edit each of these to its minimum and maximum for all three types of pitchers. I've included how I calculate each on my own and also found out how much it impacted the pitcher's overall rating.

First, to designate a player as a "veteran" I set a minimum of 200 innings for starting pitchers and 60 innings for relievers. Each is roughly a full season's worth of innings for the respective role, though admittedly they're both somewhat low. However, my reasoning is that the alternative isn't preferred but it's all I have. Players that fall under the respective limit are "rookies" and get treated with the Steamer projections on FanGraphs (I found them to be a little more forgiving than ZiPS, which can be extremely pessimistic) for some stats and also get slapped with the Joe Smith Perfectly Average treatment on others. Again, each will be mentioned when relevant.

Behold, the stats, how I calculate them, and how they affect Joe Smith and every other pitcher in the game relative to the average rating. The number in parentheses relates to how much the overall rating increases or decreases when bumping him up to a 99 or dropping to a 0, respectively.

Potential

Impact on Rating: None

Calculated: Eyeball it (pick whatever I feel is appropriate)

Reasoning: I quickly learned that changing a player's Potential does nothing. I'm not even sure if it makes a difference beyond aesthetic appeal. I did notice that when I first bought the game, the older star players (like Koji Uehara) were given A-ratings while the older mediocre players (Jamey Wright) usually had lower Potential, so I like to think that it's supposed to be like a volatility rating. Perhaps a good Potential means the player has a better chance of outperforming his ranking while a low Potential means he's more likely to underperform. I honestly don't know and this is one I've never really thought hard over. I usually give a high Potential to the stars and future stars and that's about it.

Stamina

Impact on Rating: SP (+6), RP (+3), CP (+1)

Calculated (Veterans): Innings pitched per appearance

Calculated (Rookies): Roughly high-60s to low-70s for SP, mid-10s for LOOGYs (Lefty One-Out GuYs) and ROOGYs, high-20s for one-inning relievers, mid-40s for multi-inning relievers (all estimated from Steamer projections)

Reasoning: I think using IP/app is pretty obvious. The numbers chosen for the "rookie" comes from what The Show had set for pitchers when I first started. I'm not fully sure but I think the number correspond to the number of pitches one can throw before tiring. Yes, 100 pitches is the standard in reality but this is a video game and setting all starters to 99 Stamina would hearken the game back to Old Hoss Radbourn's day when you were expected to pitch a complete game. Stamina is also less important for relievers and virtually all closers go just one inning so it makes sense that having plenty of Stamina as a closer is kind of pointless.

Pitching Clutch

Impact on Rating: SP and RP (+2), CP (+3)

Calculated (Veterans): FIP, eyeball it in between the following ranges

<3.00 FIP = 99

3.70 FIP = 50

4.00 FIP  = 40

5.00 FIP = 30

Calculated (Rookies): Steamer projected FIP, same scale as Veterans

Reasoning: Oops. Apparently I'm still using 2014's average FIP (3.74). I literally didn't know it rose to 3.96 last season, which means I guess I have some work to do when I'm done with this. When I first started 10 years ago I used ERA, but learning that FIP was more accurate in predicting future ERA was why I went with that instead. It does end up being a little unfair for sidearm/submarine and knuckleball pitchers (they can typically outperform their FIP) but you'll see that Pitching Clutch doesn't affect pitchers as much as the other stats. I have thought about using xFIP instead, but Steamer doesn't project it so I stuck with FIP.

H/9 (Hits per 9 Innings)

Impact on Rating: All (+8)

Calculated (Veterans): Batting average allowed

One stat point per .001 of batting average

<.200 AVG = 99

.250 AVG = 50

>.300 AVG = 0

Calculated (Rookies): Steamer projected batting average allowed, same scale as Veterans

Reasoning: Well, unless I were to literally use a pitcher's hits allowed per 9 innings pitched, batting average makes sense for H/9. I made .200 and .300 the cutoffs for 99 and 0 because both seemed like good numbers to differentiate between terrible and elite production.

HR/9 (Home Runs per 9 Innings)

Impact on Rating: All (+8)

Calculated (Veterans): Calculate a constant of 49.5 from .5/99 (.5 is my personal cutoff for "elite" home run prevention, divide by 99 to adjust to a 99 pt. scale)

Take the constant and divide by the pitcher's HR/9

Calculated (Rookies): Same, use Steamer projected HR/9

Reasoning: This is one I'm rather proud of creating. Last season's average HR/9 was 1.02. A HR/9 of 1 with this calculation comes out to 49.5, which rounds to Joe Smith's Perfectly Average 50. The vast majority of pitchers tend to be between 0.9 and 1.1 anyway.

K/9 (Strikeouts per 9 Innings)

Impact on Rating: All (+8)

Calculated (Veterans): Calculate a constant. Find the 10th best qualifying starting pitcher's K/9 rate, divide by 99 (to adjust to 99 pt. scale)

Take the pitcher's K/9 and divide by the constant

Calculated (Rookies): Same, use Steamer's projected K/9

Reasoning: This one is terrible. I always knew I overrated a pitcher's strikeout ability. For one, this stat was always far better than any other stat for my pitchers. Second, it's because computer-controlled pitchers were racking up 12-15 Ks per game with regularity against computer-controlled hitters. Third, I meant to use the 90th percentile qualified starting pitcher's K/9 to set the "elite" level and herp-derped my way into simply counting off the 10th-best pitcher instead. Finally, and most egregiously, this formula calculates the league average K-rate as being an 80. So yeah, I'll be going back to the drawing board on this one.

BB/9 (Walks per 9 Innings)

Impact on Rating: All (+8)

Calculated (Veterans): Calculate a constant. Find the 10th best qualifying starting pitcher's BB/9 rate, multiply by 99 (to adjust to 99 pt. scale)

Take the constant and divide by the pitcher's BB/9

Calculated (Rookies): Same, use Steamer projected BB/9

Reasoning: Like K/9, I meant to use the 90th percentile and screwed up in using the 10th best pitcher. This also slightly overrates the pitchers (MLB average comes out to 61) so I might need to tweak this a little more.

Pitch Type

Impact on Rating: N/A

Calculated (All): PitchF/X data found on Texas Leaguers website

Reasoning: Texas Leaguers isn't as robust as some pitchF/X sites out there but I prefer it for its ease

Velocity

Impact on Rating: All (about +1 per pitch)

Calculated (All): PitchF/X data found on Texas Leaguers website

Reasoning: See Pitch Type.

Control

Impact on Rating: All (about +1 per pitch)

Calculated (All): Set to 50

Reasoning: Seriously, how do you measure this? I mean, I get that this would affect a pitcher's BB/9, but we already took care of that. Besides, we have Control for each individual pitch and I have no idea how to rate each particular pitch without just making wild guesses. Thus, I just made it perfectly average. This is one where I'm open to suggestions.

Break

Impact on Rating: All (about +1 per pitch)

Calculated (Veterans): FanGraphs linear pitch values

1 stat point per .04 of pitch value

>2.00 = 99

1.00 = 75

0 = 50

-1.00 = 25

<-2.00 = 0

Calculated (Rookies): Set to 50

Reasoning: I mean, I could have taken this one literally and used the movement from pitchF/X to calculate the Break in the game. However, I went a different route and decided that Break equaled effectiveness, which is measured by the linear pitch values. In a prior post I mentioned that nearly all major leaguers are between 2 and -2 for their careers, plus it makes it easy to calculate in the stat bar.

Durability

Impact on Rating: All (+2)

Calculated (All): Eyeball it, usually set around 70-75 unless in extreme cases

Reasoning: Hear me out on this one. Pitchers are typically ready every five days, occasionally every four after a short start. Relievers can usually go two days in a row but three can render them unavailable for a day or two. I think it works, even if I keep it a little high.

Speed

Impact on Rating: None

Calculated (All): Don't really care, usually keep it around 30

Reasoning: Again, how do you calculate this? Pitchers having to run at full speed isn't exactly something they do with regularity. They definitely spend more time jogging, so I might as well keep them at a rating that suggests jogging speed.

Arm Strength, Arm Accuracy

Impact on Rating: All (+1 each)

Calculated (All): Set to 50 except for Jon Lester's arm accuracy which is at 10

Reasoning: I never used to bother with the defensive stats on pitchers. I didn't think it really mattered that much until this spring when I took note that some created relievers were far worse than others that I considered to be similar in ability. I looked at the pitching stats and saw nothing different until I explored a little further. Bingo, one was adequate on defense while the other was awful. I set both players' defensive ratings to 50 across the board and suddenly they were similar pitchers. This is another thing that really isn't measurable, and though pitchers are theoretically the hardest throwing players on the field, I decided I wanted their overall rating to be strictly a measure of their pitching ability.

You may be wondering, what's my beef with Jon Lester? It's no beef, it's just that if you attempt only one pickoff to first base in two years because you have the yips, I'm going to make sure your throwing accuracy reflects that.

Reaction, Fielding Ability

Impact on Rating: SP (+3), RP and CP (+2)

Calculated (All): Set to 50

Reasoning: Basically the same as Arm Strength and Accuracy.

I admit that I have some changes I need to make (mainly Pitching Clutch and K/9) but before I do that, here's the rankings for Twins pitchers that I currently have. They rank 26th overall in my game which sounds reasonable. Oh, and let's hope Alex Meyer figures out his control because I already have him tied as the second-best Twins starting pitcher. You'll probably argue about Nolasco's ranking but he's always had better peripherals than results, which is accounted in my methodology. Gibson's also a little low for my liking but that's thanks to a poor strikeout rate and H/9.

Starters

  1. Phil Hughes - 74
  2. Jose Berrios - 72
  3. Alex Meyer - 72
  4. Ricky Nolasco - 72
  5. Ervin Santana - 72
  6. Tyler Duffey - 70
  7. Kyle Gibson - 70
  8. Tommy Milone - 70
Relievers
  1. Glen Perkins - 82
  2. Kevin Jepsen - 77
  3. Casey Fien - 76
  4. Michael Tonkin - 73
  5. Trevor May - 72
  6. Ryan Pressly - 72
  7. Buddy Boshers - 70
  8. J.R. Graham - 69
  9. Ryan O'Rourke - 68
  10. Fernando Abad - 67
  11. Brandon Kintzler - 65
  12. Logan Darnell - 61
  13. Aaron Thompson - 57

Notes

- If a pitcher has split time between starting and relieving but is currently a reliever, I use his stats accumulated as strictly a reliever. This aids pitchers like Andrew Miller and Glen Perkins who have found far more success as relievers.

- Meanwhile in the same scenario but the pitcher is a starter, I stick with his overall stats. Typically you perform better as a reliever, so this actually gives that pitcher a slight boost rather than keeping just his .

- I always label swingmen (pitchers that regularly start and relieve) as starting pitchers. The exception is someone like Craig Breslow who was tossed two garbage time starts at the end of last year's season. He's clearly still a reliever.

- In some cases I use the Steamer projections instead of career stats. This is usually for pitchers whose recent performance vastly differs from his career work (Ernesto Frieri) or is an older (35 and up) player that may not keep up his current success (Joaquin Benoit).

- If you were curious, the top five pitchers are Clayton Kershaw and Matt Harvey (99), then Zach Britton, Sean Doolittle and Jose Fernandez (98). The worst five are Jo-Jo Reyes (48), Hector Ambriz (49), Sean O'Sullivan, Casey Coleman and Vin Mazzaro (52).

Thanks for reading if you made it this far. The post on position players will come later today.