Once a baseball leaves the pitcher’s hand, he has little control over what happens next. The batter could turn a 100-mph fastball in on his hands into a bloop single. He could blister a 110-mph line drive right at the left-fielder. Luck has always played a role in baseball, but now there are more ways to quantify it.

Sometimes, a pitcher will get extraordinarily lucky for an extended period. They may walk too many hitters and give up a ton of hard contact but maintain a strong ERA thanks to a low batting average on balls in play (BABIP). Other times, they’ll pile up strikeouts and limit walks and hard contact but still give up a lot of runs. Put Reds reliever David Hernandez firmly in the latter category.

On the surface, David Hernandez’s 5.40 ERA is lousy. Even in today’s high-scoring environment — the average ERA for relievers is 4.51 — that’s less than ideal. His FIP (2.55), xFIP (3.97), and SIERA (3.52) are well below his actual ERA. Hernandez also has a 1.2 fWAR, the highest of any Reds reliever. How is that possible? Because the peripheral stats don’t point to a below-average pitcher.

(Quick note: You’ll notice Hernandez has a -0.2 WAR via Baseball Reference. FanGraphs’ WAR equation puts more emphasis on the peripheral stats like strikeouts and walks, while Baseball Reference is more results-oriented in terms of runs allowed.)

Let’s start with the first stat many people look at after ERA: strikeouts. Hernandez is sporting a strong 29.3% strikeout rate, his highest mark since the 2012 season with the Diamondbacks when he sat at 35.3%. For reference, he was at 25.6% last season. Like many Reds pitchers, he’s using his breaking balls more often to achieve this. Hernandez has increased his slider usage from 18.9% to 31.9%, and he’s missing bats more than ever.

The slider is a relatively new pitch for Hernandez. He didn’t begin throwing it until 2017 when he was with the Angels and Diamondbacks. It’s a useful pitch because he’s able to tunnel it plays well off of the four-seam fastball. To the batter, the pitch looks like a fastball out of Hernandez’s hand and then breaks away.

The tunneling effect also helps his four-seam fastball get more whiffs, especially because it has an above-average spin rate (2,377, also slightly up from last year). A high-spin four-seamer appears to “rise” and batters often swing underneath the pitch. This effect plays up even more when a pitcher throws the heater high in the zone, which Hernandez has done all year.

As a result, Hernandez has a 35.4% whiff rate on the four-seamer, way up from 24.6% last season. That’s ninth-best in baseball among 365 pitchers who have at least 50 swings on their four-seamers.

Hernandez has also maintained a walk rate (8.3%) just below league average (8.6%). Hitters aren’t exactly squaring him up regularly, either. He has an average exit velocity of 87.0 mph, more than a full mph below the league average (88.2). He’s holding hitters to a 30.2% hard-hit rate (batted balls of 95+ mph), down from 36.5% last season. League average is 36.8%.

Why, then, is his ERA so high? The biggest culprit is, of course, sample size. With only 36.2 innings under his belt, a bad outing or two will have a more significant effect on ERA. But a lot of it comes down to the point made at the beginning of this article: rotten luck on batted balls.

Expected weighted on-base average (xwOBA) is a metric that ties all of the aforementioned factors (strikeouts, walks, exit velocity) into one clean number. In essence, it shows how a pitcher should’ve performed based on these elements. Hernandez has a .265 xwOBA, placing him in the top 9% of the league. His actual wOBA, however, is .309 — still better than average, but not in elite territory. The .044 differential is the highest among all Reds pitchers and puts him in the top 30 in baseball (min. 100 batters faced).

Hernandez currently has a ridiculously high .362 BABIP, ninth-highest among all relievers with at least 30 innings. His career BABIP is .287. Defense and the talent level of a pitcher affect BABIP to an extent. An abundance of hard contact can lead to a higher BABIP as well, which isn’t the case with Hernandez. But more often than not, a pitcher will regress to his career average. Without a significant negative change in performance elsewhere, BABIP that far above the career rate is bound to come down.

To further illustrate this, on batted balls hit less than 95 mph, Hernandez has allowed a .288 batting average. The league allows a .218 average on such balls. On balls hit at 85 mph or less, Hernandez has given up a .257; league average is .194. You get the point. Perhaps this is baseball’s way of correcting his unsustainably low .248 BABIP from 2018. But it’s not quite that simple.

The type of batted balls Hernandez allows has also contributed to the elevated BABIP. He has one of the lowest groundball rates (26.9%) in baseball. He allows a lot of flyballs and always has, but this season, he has seen a huge increase in his line-drive rate, from 20.7% to 30.1%. Line drives, of course, go for more hits than any other type of batted ball. Even line drives hit less than 85 mph still have a .616 batting average this season. Hernandez will need his line-drive rate to come down a bit for the BABIP to even out.

Still, the ball largely hasn’t bounced Hernandez’s way this season, especially over the last couple of weeks. But the underlying numbers show why David Bell continues to turn to Hernandez in important situations. If the right-hander can continue to miss bats, limit free passes, and mitigate hard contact, he should see his ERA come down the rest of the season — if fate allows it to happen, that is.

Photo Credit: Hayden Schiff. Licensing for the photo found here.

18 Responses

  1. Steve Schoenbaechler

    I agree, this is where the Saber-stats will miss information. Even some you put here.

    For example, the reason why McGuire and Pujols worked so well together. I believe they were ordered first McGuire batting then Pujols. If the pitcher got behind to McGuire, with 2 out, he was more likely not to try to finesse a breaking ball by McGuire for a strike, giving a fastball, in order to try to get McGuire out, because the pitcher didn’t want to put another runner on with a walk with Pujols coming up. And, with McGuire being a fastball hitter, that was right up McGuire’s alley.

    So, even the stats may say “simply bad luck”. But, it could very well be dependent upon the count, dependent upon who was coming up, dependent upon who is on base already, etc.

    Similar with hitters. A great hitter may have to work really hard to keep his numbers up. But, put another great hitter behind him (like McGuire and Pujols), it may be a bit easier to keep those numbers up.

    That’s why I’ve always said, would I be looking at the stats, even the in-depth stats? Sure. Of course. Anyone would be stupid not to. They are a great tool. But, they are still just that, a tool. They tell “what is happening”, not “why it is happening”. That always takes some degree of consideration, analysis, whether very small to very indepth analysis.

    And, they are rarely 100% correlation. There are people out there who would think simply because Hamilton was fast, when he is on base, he should steal the next base. Not always. Who’s catching? Who’s pitching? What’s the game status? Does Hamilton feel alright that day? Which base would he steal? So many other factors to consider.

    Like with Hernandez’s numbers. Maybe he’s getting unlucky on certain counts? Under certain game conditions? Like, are there men currently on base? In scoring positions? Which inning? How many outs? Maybe it was on certain pitches? Maybe it was to certain batters?

    Or, maybe it was just luck? Just like, in turn, managers simply use intuition at times to make certain moves, something that simply tells them, “This will work.”

    • Steve Mancuso

      You make a pretty strong claim about mis-information at the top then meander through a confusing and mistaken anecdote (at best) about two players from more than ten years ago that is completely irrelevant to the point Matt is making about the element of randomness of where struck balls land on the field.

      Hernandez has a strong strikeout rate and a lower than average walk rate. He has minimized hard contact. That is hard, measurable data. Based on the balls that were *actually hit* he has been unlucky compared to other pitchers who had balls hit the *same way.* That’s the entirety of the point.

      No one has claimed “100% correlation” whatever that even means here. But studies consistently show that metrics that control for defense and luck better predict pitcher performance than ones that don’t. Certainly the case when dealing with relief pitchers and small sample sizes.

      • Steve Schoenbaechler

        Actually, one just has to understand the game and strategy of baseball in order to understand my post. I definitely understand everything you have stated about stats. You have shown you definitely don’t understand game strategy. Or, should I say, you have shown you only understand the simplistics of the strategy of baseball. As in “what do the stats” say. You have shown you can’t get beyond the stats and explain why they occur, only that they occur.

        Simple example, it’s been obvious that Mahle has problems on full counts. Duh. But why? He doesn’t have an “out pitch”? Is it mental? He tightens up on full counts? He’s trying to get the K? He can’t hit the corners on full counts? Why? Stats don’t tell this. Once it gets analyzed further, then one can do something about it.

        Like I said, I would definitely look at the stats; they are definitely a tool. They are definitely not “the tool”. That’s something you can’t get beyond.

        Don’t worry, Steve. When we want to discuss stats, we will still give you a call.

      • Steve Mancuso

        You continue to make assertions without data to support them, even claiming things are “obvious” when they aren’t true at all, let alone obvious.

        You: “It’s been obvious that Tyler Mahle has problems on full counts.”

        The league wOBA with a 3-2 count is .377. Mahle is 10% better than league average in that count, at .342. Good ‘problem’ to have, I guess.

        No out pitch? Tell that to all the guys he’s striking out. Mahle’s strikeout rate (25%) is higher than league average (22.7%).

        Mahle has a new, more effective off-speed pitch. Here’s a post you can read to get informed about it:


    • Matt Wilkes

      I am aware that other factors play into a pitcher’s performance, but the goal of this post isn’t to break everything down to the micro level. It’s just to generally show where Hernandez has been unlucky. There are plenty of ways I could dig even deeper if time allowed it.

  2. Big Ed

    McGwire and Pujols played only one year together, in 2001. McG played 98 games and had a .808 OPS.

  3. Brian S Jolley

    Almost every time he comes in Brennaman gushes about how great he is and then he raises his ERA closer and closer to 5. Give me Jared Hughes ahead of Hernandez in most situations (unless only a strikeout will do).

    • Seadog

      I think you “totally” miss the point. Correct me if I am wrong. Give me Hernandez over Hughes anytime. This is a great article. Just proves how outdated ERA is. It has taken me a long time to understand this. ERA is important, but dig into why it is what it is. This article did that. Just awesome

  4. Jim Walker

    I am a bit of a skeptic when it comes to luck and pitchers. I certainly agree that once ball meets bat there is nothing either the pitcher or batter can control about where it goes and what happens.

    Location, movement and their command are pillars of pitching. They are within a pitcher’s control. It seems to me they are addressed in xwOBA only by inference via Exit Velocity. As Matt notes, the Hernandez line drive rate appears to be a key even though the EV of his line drives is not that great. This could suggest a slippage in command as much or more than luck.

    I think we are still waiting for a metric which might decipher such situations.

    • Big Ed

      I pretty much agree. Even the best pitchers leave a fat pitch or two out over the plate, and it usually gets hit. When a pitcher starts to go bad, due to general decline or due to just having a bad day, it typically manifests itself into his simply throwing too many fat pitches. Major league hitters clobber many of those fat pitches, and the more of them the pitcher throws, the more of them the hitters will clobber. Pedro Martinez in his prime did not throw many of them. Jason Marquis in his decline threw a lot of them.

      Sometimes we can dig too deep into ancillary stats, when the answer is more simple. In Hernandez’s case, having the line-drive percentage go from 20% to 30% is probably a good indicator that his “fat-pitch percentage” has gone up.

      Baseball never gets much more complicated than the first rule of hitting by Ted Williams: “Get a good pitch to hit.” When a pitcher serves up a lot of good pitches to hit, then that pitcher will get hit.

      • Jim Walker

        I was looking at from the other angle, pitching. One of the best parts of Tom Seaver’s tenure with the Reds was listening to him and the Ol’ Lefthander talk pitching on the Star of the Game show.

        Seaver talked constantly about A)Location, B) Movement, and C) Velocity. It was probably easy for him to say; but, he thought velocity was a distant 3rd to the other two in the order I just put them.

        Aside from the personal tragedy of his illness, I sometimes wish he was still able to offer his opinion of the current state of pitching versus hitting just to know what he thought of it.

    • ToBeDetermined


      “Location, movement and their command are pillars of pitching.” Thanks for that insight. I’ll tweak it a little if you don’t mind.
      I’d go:
      1) movement
      2) location
      3) then velocity

      • Jim Walker

        Fine by me. The order I had them was how Mr Seaver always saw it:


  5. Brian

    Not missing the point. Just making another one related to the DH subject. He is solid but nothing more.

  6. Doc

    Hernandez pounded again Sunday. I don’t see how his bad pitching is bad luck. Too many stats are excuses for poor performance in my book, stuff agents use to extract bigger contracts for worse performance.

  7. Lwblogger2

    Really good stuff. He’s been pitching better than his results. That said, at the end of the day it’s the results that lead to wins and losses. I still think he’s a decent option in higher-leverage situations compared to most the pen. That said, if the results don’t start to improve, I’d be more and more reluctant to use him in those situations no matter what the underlying metrics say.

    I love these deeper dives into pitching. They give some really good insight into the how and why parts of the game.

  8. Doc

    More bad luck in first appearance after the ASG. Two of his allowed fly balls landed in the seats. Bad luck that the architects put seats and a wall there.