The Baseball Equivalent of Hitting on 16

Fairly early in life, I’d venture to guess that many of us learned to play basic card games, from poker to rummy to blackjack. These games were often learned at home from parents or other older relatives, in a family bonding type of setting. At an early stage in this process, someone likely sat us down and handed down some helpful hints as to how to play the game well – if the dealer is showing a face card in blackjack, for instance, it might make sense to take another card – a hit – if you are holding 16, otherwise a scenario in which you would almost never take another card. Playing the game thusly doesn’t mean you’re always going to win, of course – it simply tilts the odds ever so slightly in your direction.

Whether you’re playing bridge, Scrabble, Monopoly or Trivial Pursuit, there are “little things” you can do within the rules that enhance your chances of winning. It’s called game theory, and understanding it is vital to success in any endeavor that includes an element of chance. Odds are that utilization of data has become more commonplace in your workplace, and is integral to the management of businesses of all types. For some reason, despite the proliferation of data and its increased usage in baseball today, basic tenets of game theory continue to go unheeded by managers/organizations, and unnoticed by announcers/traditional media/bloggers. Case in point – this past weekend’s Mariners-Marlins series.

The Seattle Mariners rolled into Miami last Friday in desperate need of a victory. A promising start seemed eons ago, has they had lost five of their last six games, including the final three in Texas. The most painful of those losses was a 9th-inning 3-2 loss in Texas that was a conversion of a routine grounder away from victory, wasting a masterful outing from Felix Hernandez. Their rotation was reeling from the recent losses of James Paxton and his injury replacement, Blake Beavan, to add to the previous losses of Hisashi Iwakuma and Taijuan Walker. The Marlins hadn’t been faring any better, losing eight of their previous nine games. It shaped up as your classic movable object vs. resistible force matchup. It turned out be much more than that. Perhaps a metaphor, or a how-not-to treatise on baseball game theory.

Game 1 – Friday, 2nd inning. The Marlins had just taken a 3-2 lead on a sacrifice fly by Marcell Ozuna in this matchup between Chris Young and Nathan Eovaldi. With Christian Yelich on second base and two outs, Giancarlo Stanton was intentionally walked, bringing up lefthanded hitter Garrett Jones. This move actually increased the Marlins’ win probability by 1%. Jones – with the platoon advantage, unlike Stanton – followed with an infield single, but a heads-up play by shortstop Brad Miller caught Yelich rounding the third-base bag, and the final out was recorded. An intentional walk – in the 2nd inning. This, incidentally, was Stanton’s first IBB of the season. In a vacuum, it’s a negative percentage move, but in the 2nd inning, all it’s guaranteed to do is bring Stanton to the plate sooner, and perhaps give him an extra plate appearance late in the game. File that away for later.

Fast forward to the bottom of the 4th inning. It’s now 4-2 Marlins, and Young has been relieved by lefty Joe Beimel. Yelich is again on second base, with two outs. The Mariners again issue Stanton an intentional walk, which again increases the Marlins’ win probability by 1%. This one is marginally more sensible, as Beimel at least has the platoon advantage over Jones this time, and he strikes him out. Still, it is only the 4th inning, and it again brings Stanton up to the plate that much sooner, and further enhances his chances of garnering a fifth plate appearance in the endgame. Look at it this way – in either case, the 2nd or 4th inning situation, you could simply choose to pitch around Stanton, allowing him to potentially get himself out in the process. He can only hurt you with a hit, and he has a career .267 batting average. The chances of him hurting you are roughly the chances of an NBA player missing a free throw. Execute pitches – make the darned free throw.

You know where this is headed. The Mariners gamely scrapped back with single runs in the 5th and 7th, tying it 4-4. The game headed to the bottom of the 9th, and the Mariners let middle reliever Yoervis Medina, who had recorded the final two outs of the 8th, return to the mound. Medina is a fine, functional MLB reliever – every club has one or two of him. Command is his primary weakness, as evidenced by his total of 46 walks in 76 1/3 career innings. What he isn’t is the club’s best reliever. That would be closer Fernando Rodney.

Now it must be noted that the Mariner bullpen was very overtaxed at this moment in time. They had gotten just two innings from Thursday starter Erasmo Ramirez and just three from Young on this night. The only member of the pen who had not been fully taxed at this point, however, was Rodney, who had pitched all of just 2/3 of an inning in the previous week. It wasn’t a save situation, however, so old-school baseball wisdom states that you can’t use your closer in a tie game on the road. Whatever. File this one away for later as well.

Pinch-hitter Reed Johnson leads off the home 9th with a single. The Mariners have already used both of their bullpen lefties, so they are stuck with a poor matchup of righty Medina vs. red-hot lefty Yelich. This is where the Strategy Carnival really begins on both sides. The Marlins punted their matchup advantage by bunting Yelich – who laid it down perfectly, beating the bunt out when Justin Smoak slightly bobbled it.

This brought up another red-hot hitter, Marcell Ozuna, a righty. A bunt again seemed to be an unattractive option for the Marlins – a successful bunt would almost certainly take the bat out of the hands of their best hitter, Stanton. They bunted again, and this time there was perfect execution on both sides, a great bunt by Ozuna and an even better fielding play by Medina, who barehanded it and in one motion threw it to third for the force – only for the replay gods to snatch the out away thanks to the newly minted “transfer rule”.

So here we are – bases loaded, no outs, tie ballgame, and up walks – Giancarlo Stanton. He was intentionally walked twice earlier, and as a direct result it was he specifically, and the top of the order in general that came to bat in the 9th. Even Stanton played it wrong strategically in this at-bat – all he had to do was get the ball in the air for the win. Instead, he first swung through a 1-1 hanging breaking ball in the middle of the plate, trying to hit it to Jupiter. Medina then threw the same pitch again, and Stanton didn’t miss it. At no point was Fernando Rodney warming up during the inning.

All the highlights, post-game shows and articles afterward mentioned two things – the Stanton grand slam and the immediately-preceding replay reversal. Never mind the fact that Stanton batted in the 9th solely because of suboptimal strategy by the Mariners, and notched his game-winner despite suboptimal strategy by the Marlins. If there is some justice here, the team that best executed its suboptimal strategy prevailed.

Next, on to Saturday night, and the least eventful contest of the three-game set, a 7-0 Marlin victory, a two-hit shutout by Henderson Alvarez. It was notable from a suboptimal strategy standpoint for two reasons, however. First, there’s the handling of Mariner starter Roenis Elias. The young Cuban lefty had never pitched above AA entering this season, but the lefty earned a spot in the Mariner rotation thanks to a strong spring performance coupled with the attrition surrounding him. He pitched competently in his first three outings, but was fighting for each out on this night, as he battled into the 6th inning down only 2-0.

As noted previously, the Mariner bullpen was in tatters at this point, so they tried to squeeze one more inning out of him. As his pitch count climbed toward a career high 111, he faced a runner on 2nd, two out situation, with only opposing hurler Alvarez standing between him and the end of a respectable outing. Alvarez touched him for a single to make it 3-0. That had to be it for Elias, right? Nope. He then faced lefty Yelich – and walked him. Pull him now, right? Nope. He was allowed to face righty Ozuna, who drilled a three-run homer. Now, Danny Farquhar comes in. Oh, and in the 8th, guess who came in to get an inning of work? None other than closer Fernando Rodney. A world in which he can’t come into a 4-4 game one night but must come into a game with a 6-0 deficit the next night is a nonsensical world, indeed.

This article would not have been written without what happened in Game 3, however. The recent injury to injury replacement Blake Beavan forced the Mariners to call up righty Brandon Maurer from AAA to start, and he was brilliant for 4 1/3 innings, facing the minimum 12 batters through four, and leaving with a 2-1 lead. He faced the mighty Stanton twice, inducing a double-play grounder and striking him out. Rookie reliever Dominic Leone followed suit in the 6th, striking him out. Make quality pitches, hit the free throw, get the out.

The game remained 2-1 Mariners into the home 8th, when the final act unfolded. The ever-present Yelich – told you he was a good fantasy draft, Jessica – led off with a double off of lefty Charlie Furbush, who was then replaced by righty Tom Wilhelmsen, who promptly got Ozuna to fly out. Up walks Stanton, this time representing the winning run. The same Stanton who had been utterly neutralized by less experienced and pedigreed hurlers the previous three times up. The Mariners walked him intentionally, this time increasing the Marlins’ win probability by 5%. They didn’t try to make the free throw. They took a hit on 16.

Of course, the next batter, Casey McGehee, was unintentionally walked to load the bases. Such a result has a higher chance of happening than the DP grounder the Mariners were chasing. The rest was a mere formality. Yelich was safe at the plate on a fielder’s choice grounder – but only after a replay reversal – to tie it, and Adeiny Hechavarria followed with a sacrifice fly to win it. All that remained was for Marlins’ closer Steven Cishek – who did pitch in the tie game on Friday – to shakily fight through a man on third, one-out jam in the 9th for the save.

This article is not meant to pick on Lloyd McClendon or the Mariners in general. It could have been any number of managers or clubs – it just happened to be this one that did all of these things repeatedly in a single weekend, and lost more than one game as a direct result. Managers have the hardest job in baseball, in my opinion, and game strategy comprises a very small percentage of it. Managers are hired to be leaders of men, who are in this case often millionaires many times over, and are expected to hold their attention and respect for a long, 162-game marathon. I would argue that it is the responsibility of the organization to educate their field personnel about game theory, about the math behind the usage or non-usage of various strategies.

This is obviously taking place throughout the game with regard to infield overshifting, and field staff are obviously responding to the data being shown to them by implementing such strategies. This isn’t a one-way street – field staff obviously must have some degree of autonomy, and need to have a voice in development of on-field plans, but a front office that doesn’t share and effectively communicate data that can give their club a competitive advantage – or at least avoids placing them at a competitive disadvantage – does so at its own peril.

A healthy interaction between front office and field staff regarding implementation of data helps to forge a bond between them that is becoming increasingly vital. If it is the field staff’s responsibility to be open to all kinds of new information, it is the front office’s responsibility to have the field staff’s back on the inevitable occasions when that openness to good process leads to bad results. In the big picture, each front office has a responsibility, not only to its field staff, but also to its players and all of its core constituencies, including its fanbase, to maximize its organizational IQ.

When you think about it, these lingering suboptimal strategies are seemingly limited to baseball these days. In no other sport will you see a lesser player purposely entrusted with a central role in a game situation at the expense of an admittedly better player. In no other sport is a game strategy repeatedly undertaken that has a measurably negative effect on that team’s chances of winning. In no other sport will a team consciously forego attempting a task that has virtually the same probability of success as a free throw attempt. These things continue to happen every day in our sport.

There is a time and place for an intentional walk – it’s usually in front of an opposing pitcher, or a very weak hitter who isn’t likely to be replaced by a pinch-hitter. There is a time and place to walk Giancarlo Stanton – and that exact spot might have presented itself in the 9th inning on Friday night if a base was open. There’s a time and place for a sacrifice bunt – usually with a pitcher or weak hitter, or in an extreme low-run environment, late in a close game. You do have to pick your spots on when to use your best reliever – he only has so many bullets. If he’s barely pitched in a week, and the game is tied in the 9th, you just might want to get him in there.

There is hope, and progress, as AL sacrifice bunts are down 25% per plate appearance since just last year, and are over 40% down since 2010. NL intentional walks are down over 20% per plate appearance since just last year, and are over 40% down since 2006. For the math behind the relative efficacy of these situations and more, “The Book” by Tom Tango is a must read. He has every loose end tied up tight.

Similarly, there’s a time and place to hit on 16, but if you’re doing it regularly and without regard for context, you’re going to get burned. You’re going to become “that guy” in the card game, the guy no one wants to sit beside. At this stage of the game, with the preponderance of data at everyone’s disposal, we should be beyond this. No single team’s fanbase should have to wake up on Monday morning and stare at the ashes of a three-game sweep that shouldn’t have happened, that was largely attributable to bad process, to suboptimal decision-making that could have easily been avoided.

Print This Post

87 Responses to “The Baseball Equivalent of Hitting on 16”

You can follow any responses to this entry through the RSS 2.0 feed.
  1. tz says:

    Cliff Notes version:

    Well-Known Secondary Effect in Blackjack:
    When you hit on 16, it brings up the other cards in the deck one card sooner.

    Less Well-Known Secondary Effect in Baseball:
    When you intentionally walk the opposing team’s top slugger early in the game, it brings him up to the plate

    +33 Vote -1 Vote +1

    • tz says:

      …one at bat earlier.

      +5 Vote -1 Vote +1

    • Richie says:

      True Secondary Non-Effect in Blackjack:

      When you hit on 16, it brings up the other cards in the deck one card sooner. And if you think this matters one whit, your grasp of probabilities is not quite as great as Lloyd McClendon’s.

      +15 Vote -1 Vote +1

      • tz says:

        Card-counters or wannabe card counters would beg to differ. But aside from that, you are totally right.

        +5 Vote -1 Vote +1

        • Corey says:

          You bring up a good point, so if I’m screwing around and (stupidly) hit on 16 and the guy next to me throws a temper tantrum I should ask him if he’s counting cards?

          Vote -1 Vote +1

        • Cozar says:

          @ Corey

          Other way around. If you are screwing around and stupidly hit on 16 and the guy next to you DOESN’T throw a tantrum, then you should ask if he is counting cards.

          The default position is don’t hit on 16, the exception is if you are counting cards and know there are better-than-normal odds of not busting by hitting.

          Vote -1 Vote +1

        • Corey says:

          It doesn’t matter if I hit on 16 unless you’re counting cards because you don’t know what the next card is, your odds are completely unchanged, I could just as easily have taken the shitty card from you as the good card.

          Vote -1 Vote +1

        • Cozar says:

          Really? The odds don’t change? Assume two players + dealer, none of the three cards showing are a ten or face card. The odds of player number #2 drawing a card valued at less than 10 are 33/49. If player #1 hits on 16, those odds change to either 33/48 or 32/48.

          The odds change. Unless the player is counting cards, he/she wont have enough information for your poor decision to effect their strategy, but that doesn’t change the fact you did change their odds.

          Vote -1 Vote +1

        • …those odds change to either 33/48 or 32/48

          If the odds change, but I can’t tell (b/c 6 card deck is on constant shuffle) does it matter?

          Vote -1 Vote +1

      • Eric R says:

        My first time playing blackjack in a casino with other people at the table, the first time I hit when “Black Jack Optimal Strategy” said I shouldn’t the guy next to me grumbled something about me “taking his card”.

        The next time it happened he got up and left, but told me to sit on the left side of the table if I’m not playing optimal strategy as not to “take” other people’s cards [or give them “my” card if I stay when I should hit].

        Vote -1 Vote +1

        • KK-Swizzle says:

          FYI you aren’t actually affecting his odds at all. His response is a more or less an offshoot of gambler’s fallacy and should be disregarded.

          That being said, you shouldn’t ever be hitting 16 against a dealer up card of 6 or less unless you are trying to lose money :)

          +10 Vote -1 Vote +1

        • MH says:

          Actually, you are affecting his odds (if he’s a card counter)–in a positive way. Every card drawn from the shoe gives a card counter more information about the cards remaining. Objectively, and in general, (this probably isn’t true for each specific situation, though I’m not certain enough about the theory) card counters should want to play at tables with players who hit too much because they’ll be given the benefit of added data for every decision they make. The only downside is they run the shoe faster. But all that means to a disciplined card counter (who has lots of free time) is that they can go watch another table for a bit and sit down again after feeling comfortable with the shoe.

          Vote -1 Vote +1

      • Roger Clemens says:

        I’m glad this was about blackjack. Got scared when I saw “Hitting on 16″

        +44 Vote -1 Vote +1

      • Bip says:

        yeah, i don’t get it, it’s not like you know what any of those cards are anyway.

        Vote -1 Vote +1

  2. Jason B says:

    Awesome artcle Tony. Really good read. I think you hit on the crux of the issue late in the article – that the front office needs to pass along the data to the manager with the expectation that it will be communicated and implemented prudently, and then stand behind them in support when a non-traditional move inevitably falters. As they will, often. Just not quite as often as the traditional move will.

    Vote -1 Vote +1

  3. AC of DC says:

    Good stuff, though I cannot agree that only in baseball are these strategic errors executed. Basketball, football, soccer, and hockey all feature input from the coaches that can call for bad matchups and assignments, the wrong play for the situation, and the wrong guy for the job. The more fluid nature of the play in these sports precludes the discrete exchange that is featured in baseball, so the mistake isn’t necessarily clear as quickly, especially to a casual viewer.

    +7 Vote -1 Vote +1

  4. Kris says:

    Football coaches routinely pass up optimal game strategies and are so risk averse it’s sometimes hard to watch. I doubt that baseball is the only sport with lingering suboptimal strategy.

    Amazing article, and fun to read.

    +36 Vote -1 Vote +1

    • Yes, this. Football is even more suboptimal than baseball. The amount of risk aversion, and what frankly appears to be anti-intellectualism in the sport is pretty rampant.

      Good article though. Front offices have come a long way (even though they definitely aren’t there yet), but the guys in charge on the field have a long way to go to catch up.

      +22 Vote -1 Vote +1

      • tz says:

        Any coach punting on 4th-and-1 at midfield when down 2 or more touchdowns in the 4th quarter deserves to be shot right on the spot. Unless his team is trying for the #1 draft pick.

        +17 Vote -1 Vote +1

      • Rick says:

        Suboptimal strategy in terms of winning a game can lead to a more optimal chance of keeping a job for a coach/manager. It all depends on what is more important to the coach/manager.

        +11 Vote -1 Vote +1

      • mike wants wins says:

        Agreed, NFL coaches are universally afraid of ridicule, often punting on 4th and short. Check out the many articles by Gregg Easterbrooke on ESPN on the topic.

        +8 Vote -1 Vote +1

      • DL80 says:

        Agreed. The fact that almost no coach ever goes for it on 4th and 1 is a pretty good analogue. I’m also not sure whether the “Hack-a-Shaq” strategy in basketball is a net positive or negative in terms of likelihood of winning.

        And what about having your worst shooter/scorer go first in a shootout in hockey or soccer?

        Vote -1 Vote +1

        • Ruki Motomiya says:

          I think that the thing with 4th and 1 is pretty overstated: The problem is teams not going for it when they have a shitty defense, b ut if you’re a team like the Seattle Seahawks or San Francisco 49ers then turning it over to your D is a good call. In addition, you have to consider quality of opposing offense (It is better to go for it against a worse offense and worse to go for it against a better one) and field position (Going for it closer to your own end zone is riskier).

          Vote -1 Vote +1

      • John Elway says:

        Hay, don’t even get me started on the “prevent” defense. Even if you have ancient relics in your defensive backfield, there’s just no excuse for letting the other team march downfield quickly late in the game.

        +14 Vote -1 Vote +1

        • Ruki Motomiya says:

          If you’re up like 4 TDs with 5 minutes to go than it’s a viable strategy, the Prevent is just turned too with leads that are way too slim.

          Vote -1 Vote +1

      • Iron says:

        I recommend the ‘Football Outsiders’ website to Fangraphs readers who want a site with a similar approach to football.

        Vote -1 Vote +1

      • Mr Baseball says:

        the NFL is a joke when it comes to optimal strategy. They allow the fan base, media and conventional wisdon to dictate. The NFL is so far behind Texas HS football it’s not even funny. The fact that the 2pt conversion is on the two yard line, not three, as in colllege, is reason enough to attempt a 2pt conversion much more than they do already…but no.

        Nevermind all of the other crap NFL coaches do that we are all accustomed to.

        Vote -1 Vote +1

  5. Warning Track Power says:

    Well done analysis, though I would argue the home team bunting in a tie game in the ninth is defensible, if not necessarily always optimal.

    Vote -1 Vote +1

  6. Fred says:

    So, as I understand it, you got sacked from the Mariners because they wouldn’t listen to your presentations using advanced data. And now, you can write a blog post going into great detail about how organizations who don’t teach their managers and players how to use advanced data get burned sooner or later.

    I’m sure the schadenfreude you’re having over the state of the Mariners is pretty fun, Tony.

    And I don’t mean any of that as a swipe at you. The exact opposite, actually. You’re justified. You and many others have laid out all kinds of stats from all kinds of situations where the M’s ignored some pretty straightforward information.

    +13 Vote -1 Vote +1

    • Ron says:

      I actually wish this wasn’t about the Mariners and that he had waited till another team did something similar. While the information presented was informative and very good, I don’t think this post will get the credibility it deserves because of the author and team it is about.

      Vote -1 Vote +1

  7. clint h says:

    is the author aware that it is correct to stand on 16 against dealer’s 10 more often than not?

    hit on 16″ is blackjack folk wisdom that knowledgeable players ignore.

    how was the change in win percentage for walking stanton calculated? it seems silly to quote his lifetime average since he is both very young and has often played hurt. i cannot imagine not walking him intentionally when the next card in the deck is post-prime garret jones.

    Vote -1 Vote +1

    • Tre says:

      No, it’s not. Think about it using math.
      2-5 cards (the 2,3,4,5) are 16 cards in total. The other 32 cards are seven or higher (not including the 4 sixes). We know that a six, and two face cards are out there. So that means that there are 30 cards that could be the face down card that beat your 16. 30/52 (or ~58%) is the probability you will lose by not hitting.

      Vote -1 Vote +1

      • Tre says:

        Actually 30/49 because you know three cards ~61%

        Vote -1 Vote +1

        • Thor says:

          Umm, am I missing something? Yes, there is a 61% chance that the dealer’s face down card can beat you. But if you hit, there is a 36/52 (69%) chance that you will get a card that takes you over 21, and you will lose no matter what the dealer has.

          Vote -1 Vote +1

        • Steve says:

          And the odds you lose by drawing a card greater than 5 are?

          Vote -1 Vote +1

        • Rick says:

          61% chance of losing if you don’t hit isn’t exactly correct, b/c dealer could have less than 7 underneath and still hit for total of 21 or lower. Chance of losing if you don’t hit are closer to 70% than 60%.
          Yet if you hit, 59% chance of busting, plus another ~10% of the time you hit w/o busting and dealer still has a higher total less than 21 = ~70% chance of losing if you do hit.
          End result, having 16 when dealer showing a face means you lose 7/10 no matter what strategy you use (unless you’re counting cards).

          Vote -1 Vote +1

        • Tre says:

          Nvm, there is a 61% chance that you are currently losing. But you guys are correct by hitting there is only a small chance of actually improving your hand and not going bust.

          Vote -1 Vote +1

      • m_pemulis says:

        You are making the colossal mistake of assuming that because you have a >50% chance of losing by standing, that the alternative (hitting) is a better move. In fact, the 50% threshold is meaningless (some starting blackjack positions are inherently more disadvantageous than others).

        Consider what happens when you hit. 33/49 (67%) cards will bust you, resulting in an automatic loss. That alone shows you have a higher probability of losing by hitting PLUS there is the added probability that you get a 2 or 3 and the dealer is holding 20 (and other permutations of the same idea) and you would lose anyway.

        The main takeaway is that when you see a 16 in your hand and a face card showing for the dealer, you are already in a shitty situation and will lose over 50% of the time. However, by not hitting, you put yourself in a slightly less shitty situation (lesser of two evils). The same goes for making suboptimal decisions when there are runners in scoring positions and 0 out. Sure, the chances are you aren’t gonna get out of it with no runs scored on you, but you should still try to limit the damage.

        Vote -1 Vote +1

    • Los says:

      The only time you should hold on 16 is with an extremely high count in a thin deck. However, your odds are even worse in those situations than hitting on a low or neutral count.

      Vote -1 Vote +1

    • KK-Swizzle says:

      The correct play is actually to surrender that hand if allowed :)

      Vote -1 Vote +1

    • Spencer says:

      Just like in the movie Wargames, the correct move is to not play.

      Vote -1 Vote +1

    • Cragly says:

      If player has 16 and dealer has 10, your expected return:
      Stand: -0.54
      Hit: -0.53
      Surrender: -0.50

      The two are close enough that the count can quickly make a difference. If the count is positive, you should typically stand.

      Vote -1 Vote +1

  8. Nate says:

    Of course, the next batter, Casey McGehee, was unintentionally walked to load the bases…

    Just for giggles, this was the MLB Gameday image of that “unintentional walk”.

    Vote -1 Vote +1

  9. diderot says:

    Fred is correct. Blengino is now doubling down on his decision to burn bridges not only with the Mariners, but with everyone else in MLB. Hope Cameron’s paying you enough, because this is now your career.
    Too bad you can’t get paid for doing nothing, like your pall Geoff Baker.

    -10 Vote -1 Vote +1

  10. chuckb says:

    I was eating lunch at my desk while reading this and when I got to the part about Rodney entering the blowout game I laughed so hard I literally spit food onto my desk.

    Great stuff, as always, Tony. Remind me not to read your articles while I eat, however.

    Vote -1 Vote +1

  11. JB says:


    Vote -1 Vote +1

  12. Dustin says:

    McClendon faced a situation during the 9th inning of last Wednesday night’s game in Texas which was also perplexing. Shortly after the Miller error, Rodney threw a wild pitch with Leonys Martin batting which allowed the tying run to score. The count reached 2-0 with a runner on second and the winning run on third. Rather than walk Martin with the dreadful JP Arencibia on deck, McClendon elected to pitch to Martin. Martin then singled in the winning run. The move did not make much sense to me. First base was open and Martin was now in a hitter’s count. McClendon could have faced a much more inferior hitter who has contact issues with a force at any base. This is obviously a very small sample, but McClendon and his staff appear to have issues maximizing their team’s chances of winning.

    Vote -1 Vote +1

  13. SucramRenrut says:

    Also Lloyd McLendon has always been a terrible manager in the Dusty Baker, Ron Washington, ‘I plays bazeball so I managez’ line. Still can’t believe anyone would hire him.

    Vote -1 Vote +1

    • Ron says:

      I don’t know how many managers Jack Z had to really chose from. I know MLB jobs are rare and people want them, but look at Jack’s level of success? If you were looking for a job and it looked like your boss might get fired and the new boss might want to bring in someone else, would you take that job?

      Vote -1 Vote +1

      • Corey says:

        As a Mariners fan I was hoping they would hire someone with nothing to do with MLB who understands odds. Wishful thinking regardless of GM I think.

        Vote -1 Vote +1

  14. Grohman says:

    “If there is some justice here, the team that best executed its suboptimal strategy prevailed.”

    FanGraphs Quote of the Day.

    Vote -1 Vote +1

  15. Atreyu Jones says:

    “Similarly, there’s a time and place to hit on 16, but if you’re doing it regularly and without regard for context, you’re going to get burned. You’re going to become “that guy” in the card game, the guy no one wants to sit beside.”

    In an article about logic and game theory, why give credence to the superstition that sitting beside a sub-optimal blackjack player affects your odds?

    +5 Vote -1 Vote +1

  16. Pirates Hurdles says:

    “Never mind the fact that Stanton batted in the 9th solely because of suboptimal strategy by the Mariners”

    This isn’t necessarily true (and thus not fact), Stanton could have reached base in his previous at bats anyway. He could have gotten hits that changed the game situation completely. Sure its suboptimal by percentages, but it doesn’t mean the outcome is known. I tend to agree with the conclusions in The Book, but lets not act like situational events have no bearing on outcome.

    +7 Vote -1 Vote +1

  17. Carl G says:

    Do you think for one second that the butterfly effect doesnt apply here? You need to re-evaluate your stuff. I know its hard to come up with great polarizing columns every week, or few days, but if you pitch to the batter instead of walking him, the next pitch thrown is a completely different strategy. Im sorry, this column was a waste…creative, but false. Im not a hater either, I love your site, have read, and will continue to read all of your stuff, but Im sorry, I have to call a spade, a spade.

    Vote -1 Vote +1

    • Aaron says:

      If you examine all possible butterfly effects, there are more likely outcomes in favor of pitching to Stanton than to pitch around him. I remember when the Indians had the same positive results when players would pitch around or walk Albert Belle. That extra, high leverage, plate appearance later in the game helps his team plus puts him on base in the early innings. Just pitch to him.

      Vote -1 Vote +1

  18. Corey says:

    Wow! This is a long comment thread!

    It’s not really true that this only happens in baseball. My pet peeve in football is teams punting in opposing territory on 4th and short. What percentage of the time do they expect to gain 2+ yards and how valuable is it to put your opponent on their own 20 instead of their own 40?

    I think the real point here however is, as the Marlins engaged in stupid strategies too (bunting), that the Mariners can’t seem to execute at all right now, they must lead baseball in little mistakes that end up hurting big. That said, they might get away with some of their mistakes if they weren’t pursuing suboptimal strategies the mistakes probably wouldn’t hurt them as often.

    Vote -1 Vote +1

    • Ruki Motomiya says:

      I think the big thing with punting is it depends on the team: For a team like the Denver Broncos it is really stupid to punt in that situation, but for a team like the Seattle Seahawks the potential gains on defense such as field position from sacks, turnovers, or just holding them and getting it back, makes a large difference. Plus, an ideal punt around then will put the foe at their own 10 or so, which is very dangerous territory for stuff like sacks and limits playcalling options.

      Vote -1 Vote +1

  19. Joe says:

    I don’t entirely agree with all that you say. My biggest problem is your notion that sacrifice bunting in the 9th is the suboptimal strategy for the Marlins. Sacrifice bunting necessarily lowers your odds of scoring multiple runs in an inning, but I think it could potentially increase your odds of scoring a single run. And obviously in the bottom of the 9th in a tie ball game, the Marlins only need to score one. I would like to see the run expectancy numbers for a runner on 2nd with one out vs. runner on 1st no outs and again bases loaded one out (assuming Stanton is intentionally walked) vs. 1st and 2nd no outs. I understand these aren’t the most reliable data, but I would think that moving a runner into scoring position or onto third base with less than 2 outs may increase your chances of scoring a single run which obviously is the most important factor here. It’s very possible that it would still turn out that bunting is the suboptimal strategy, but I would’ve been interested in more data rather than just stating it as a fact. This is just a thought I don’t have any exact numbers or anything but what do you think?

    Vote -1 Vote +1

  20. Brian says:

    Great article!

    Vote -1 Vote +1

  21. mwickham25 says:


    +13 Vote -1 Vote +1

  22. Blackjack says:

    Anyone else reading this and think it just sound like a disgruntled former employee (Tony) trying to tarnish his former employer (Mariners). Yes they used sub-optimal strategy, but using what DID happen as proof that it was wrong seems like a poor way to argue. Hindsight is 20/20, and had any of these plays gone the other way (say a strike out of Stanton or a DP), they would’ve looked smart, despite the fact that it was still sub-optimal. The Mariners moves where not the smartest, but they were also not as bad as Mr. Blangino is making them out to be, instead relying somewhere in the middle. You are quiet a smart fellow, and despite my dislike for your style of writing, your articles are often insightful, and I’m hoping that your next article will be more based in substance rather than relying on game results to prove a point.

    Vote -1 Vote +1

  23. Tim says:

    This is a perfectly reasonable, if clumsily expressed, set of tactical points. However, it has absolutely nothing to do with the mathematical field of game theory. Open Yale has a nice intro to GT course if you’re interested in educating yourself.

    Vote -1 Vote +1

  24. Mr Baseball says:

    Yea, I do agree dropping “game theory” is an attempt to seem more intelligent than he might be. He wasn’t able to really come up with any examples that tied baseball to game theory. In baseball the best examples of game theory would be pitching away from a hitter’s strength and the hitter countering. Or if hitters actively hit against the shift…which hasn’t happened just yet.

    Vote -1 Vote +1

  25. Satoshi Nakamoto says:

    I can’t help but think something is wrong with this article’s discussion of the intentional Stanton walks. Coming into the series Stanton had 5 HRs and 21 RBIs with .299BA and .944OPS. He was killing the ball. Regardless of pitcher/batter matchups to follow it doesn’t seem bad to walk him.

    But a bigger problem with this article’s line of reasoning is it assumes that because of the IBBs Stanton will automatically get to bat sooner, or at all in the late innings. Or that if he does come to bat sooner in late innings it will even matter.

    The moment SEA chose to IBB him the first time, every single pitch and at bat afterwards was totally changed. Anything could have happened. SEA could have gotten out of the inning unscathed and then based on that positive momentum their own hitters might have batted around the following inning and put the game out of reach.

    Unless this article states that its dealing completely in 20/20 retrospective theory and not arguing to alter future decisions it seems fundamentally flawed.

    Vote -1 Vote +1

    • American Stephen Crane says:

      The probability of Stanton getting another at-bat increases with an IBB relative to pitching to him. That seems pretty straightforward and subsequent events don’t change that probability (subsequent events produce new probabilities).

      Vote -1 Vote +1

      • Dan says:

        Sure, but it doesn’t mean that future at bat will be in a higher leverage situation.

        Vote -1 Vote +1

      • Satoshi Nakamoto says:

        every at bat and play after the first IBB will be changed by the decision to ibb.
        Stanton might not get a 9th inning AB at all, or it might not matter because SEA could have scored 20 runs. I’m saying you can’t automatically fault SEA for he IBB. At that moment it made sense.
        Going back over a finished game and saying those are the mistakes that directly led to losing the game isn’t constructive.

        Vote -1 Vote +1

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>