Search the Community

Yes, the number in the title is not an exaggeration. The true odds of the moment we find ourselves in at the close of the Twins 2020 postseason campaign, brief as it was, comes to one in 69 billion. Somewhere in the galaxy, Zaphod Beeblebrox fired up the Improbability drive and Minnesota got caught in the wake. Let's do some math, shall we? Setting a baseline Now, there's a very good chance you've seen the number 262,144 floating around Twins Twitter in the last day or two, and that's because if you were to flip a coin 18 times, the odds of each flip resulting in the same outcome are 262,144:1 against, or 218. Already, this feels bad. This feels unfair. We want to fight against this statistic. BASEBALL GAMES AREN'T COIN FLIPS, I hear you cry out. So many of those games were as underdogs against the almighty Yankees, surely the odds weren't THAT bad? And yeah, from that perspective, you'd be correct. @Awoodruff3 on Twitter looked at the problem from a gambling odds perspective: https://twitter.com/awoodruff3/status/1311415416456085510 28,524:1 against! Already, this is 10 times as likely as the coin flip scenario, so the sting should only be a fraction of what we currently feel, right? Sadly, no. Here's how it really breaks down. The Methodology I have gone into the Fangraphs archives for each of the 18 games in the losing streak and made note of the moment in time where the Twins had the highest expected win probability. In 17 of the 18 games, the Twins were favored to win - and in a few cases, extremely favored - before eventually taking the L. With that information, we can look at the odds of losing from these moments where the Twins had the greatest amount of leverage to create a future other than the ones we find ourselves in now. October 6th, 2004: ALDS Game 2 [table] Game State Bottom 12th, 1 out, Twins ahead by 1 at New York Twins Win Probability 87.3% How'd things look? With Joe Nathan on the mound for his third inning of work, John Olerud strikes out and the Twins are 2 outs away from taking a 2-0 lead in the series. What happened? Nathan gives consecutive walks before A-Rod hits a ground rule double, followed by an intentional walk and a Matsui sac fly to win. Odds of a loss 7.87:1 against[/table] October 8th, 2004: ALDS Game 3 [table] Game State Top 2nd, 2 outs, Twins ahead by 1 vs. New York Twins Win Probability 64.5% How'd things look? Carlos Silva gets a ground ball out from Bernie Williams. It's still early, but teams in the lead tend to stay in the lead. What happened? Silva immediately gave up 5 consecutive singles and 3 runs before the 2nd inning was over, and the Twins never saw daylight again. Odds of a 2-game losing streak 22:1 against[/table] October 9th, 2004: ALDS Game 4 [table] Game State Bottom 7, 0 outs, Twins ahead by 4 vs. New York Twins Win Probability 97.0%(!) How'd things look? A-Rod concludes a 1-2-3 inning in the top of the 7th by fouling out to first base. The Twins have retired 9 straight batters. What happened? Yankees tie the game in the top of the 8th on an RBI single and a 3-run homer, game goes to extra innings, Yankees take the lead in the 11th, Twins fans begin to wonder if this is the start of something dire. (Narrator: It is.) Odds of a 3-game losing streak 739:1 against[/table] October 3rd, 2006: ALDS Game 1 [table] Game State Bottom 1, 0 outs, tie game vs. Oakland Twins Win Probability 58.3% How'd things look? Luis Castillo leads off for the Twins with a walk. This would be as good as it got. What happened? Frank Thomas homers to take the lead in the 2nd, and despite the Twins making things interesting in the bottom of the 8th, they would never be favored again. Odds of a 4-game losing streak 1,773:1 against[/table] October 4th, 2006: ALDS Game 2 [table] Game State Bottom 6, 0 outs, tie game vs. Oakland Twins Win Probability 57.6% How'd things look? Down 2, the Twins start the bottom of the 6th with consecutive homers by Cuddy and Morneau to tie the game, and Oakland goes to the bullpen. What happened? Oakland responds in the 7th with 2 runs off an inside the park home run. Odds of a 5-game losing streak 4,181:1 against[/table] October 6th, 2006: ALDS Game 3 [table] Game State Top 2, 0 outs, tie game at Oakland Twins Win Probability 56.3% How'd things look? Morneau opens the 2nd with a double, with Torii Hunter on deck. What happened? Morneau doesn't score, Oakland opens up a 4-0 lead, and eventually win 8-3. Odds of a 6-game losing streak 9,569:1 against[/table] October 7th, 2009: ALDS Game 1 [table] Game State Top 3, 2 outs, Twins ahead by 2 at New York Twins Win Probability 68.7% How'd things look? Twins take a 2-0 lead when Joe Mauer scores on a Jorge Posada passed ball What happened? Yankees immediately tie the game on a Derek Jeter home run, and the Twins never score again. Yankees win 7-2. Odds of a 7-game losing streak 30,571:1 against[/table] October 9th, 2009: ALDS Game 2 [table] Game State Top 9, 0 outs, Twins ahead by 2 at New York Twins Win Probability 91.7% How'd things look? After the Twins take a 2 run lead in the top of the 8th, the Yankees go down 1-2-3, and Joe Mauer comes to the plate to open the 9th. What happened? Yankees tie it up in the bottom of the 9th, Joe Mauer hits a double in the 11th that Phil Cuzzi incorrectly rules foul, and settles for a single- only to be followed by 2 consecutive singles that would have scored him had the double stood. Instead, Mauer doesn't score, Yankees walk it off on a Mark Teixeira homer, and just typing out this sentence makes me want to die inside. Odds of an 8-game losing streak 368,333:1 against[/table] October 11th, 2009: ALDS Game 3 [table] Game State Top 7, 1 out, Twins ahead by 1 vs. New York Twins Win Probability 72.6% How'd things look? The Twins have struck first on an RBI single by Mauer, and the Yankees have responded with a Mark Teixeira groundout. What happened? Yankees immediately take the lead with home runs by A-Rod and Posada. Twins threaten to tie in the 8th with a leadoff Punto double, but fail to capitalize. Yankees win 4-1. Odds of a 9-game losing streak 1,344,281:1 against[/table] October 6th, 2010: ALDS Game 1 [table] Game State Top 6, 1 out, Twins ahead by 3 vs. New York Twins Win Probability 87.7% How'd things look? Francisco Liriano has only given up 2 hits to the Yankees, who are down three and open the 6th with a Nick Swisher strikeout. What happened? The wheels come off moments later as Lirano gives up a double, a wild pitch, 2 singles, and a triple to give the Yankees a 4-3 lead. The Twins would later tie it, only to lose 6-4. Odds of a 10-game losing streak 10,929,120:1 against[/table] October 7th, 2010: ALDS Game 2 [table] Game State Bottom 3, 0 outs, Twins ahead by 1 vs. New York Twins Win Probability 66.5% How'd things look? Twins opened the scoring in the 2nd on a Danny Valencia sac fly, and the Yankees go down 1-2-3 in response. What happened? Yankees would later take a 2-1 lead before Orlando Hudson ties the game with a solo shot, but that tie doesn't last long. Yankees win 5-2. Odds of an 11-game losing streak 32,624,240:1 against[/table] October 9th, 2010: ALDS Game 3 [table] Game State Top 2, 0 outs, Twins tied at New York Twins Win Probability 50% How'd things look? This is the only game in the 18-game streak where the Twins were never favored. It remained 50/50 after both teams failed to accomplish anything in the first inning. What happened? Twins fall behind in the 2nd, and never get close, losing 6-1. Odds of a 12-game losing streak 65,248,481:1 against[/table] October 3th, 2017: AL Wild Card [table] Game State Top 1, 1 out, Twins ahead by 3 at New York Twins Win Probability 81.8% How'd things look? You remember this inning, right? Twins go into Yankee Stadium and immediately knock Luis Severino out of the game with homers by Brian Dozier and Eddie Rosario, followed by an Escobar single and a Kepler double. 3 run lead, 2 men on, only 1 out. We've got this. Yankees don't have a CHANCE. What happened? Buxton and Castro strike out to end the inning, Ervin Santana gives up the lead on a 3-run homer, Yankees win 8-4, and everyone in my generation starts to develop serious anxiety complexes revolving around who the hell we hurt to cause this. Odds of a 13-game losing streak 358,508,138:1 against[/table] October 4th, 2019: ALDS Game 1 [table] Game State Top 3, 2 out, Twins ahead by 2 at New York Twins Win Probability 67.1% How'd things look? Twins were already leading 1-0 when Nelson Cruz comes up big with a solo home run against James Paxton. What happened? As per usual, Twins lose the lead immediately. They tie things up in the 5th, but that also doesn't last. Twins lose 10-4. Odds of a 14-game losing streak 1,089,690,390:1 against[/table] October 5th, 2019: ALDS Game 2 [table] Game State Top 1, 1 out, Twins tied at New York Twins Win Probability 53.6% How'd things look? Inexplicably known as the Randy Dobnak game, the Twins were statistically favored for the briefest of moments when a HBP and a single put 2 men on in the first inning with only one out. What happened? Those baserunners are stranded on a double play, Yankees score first and never look back. Twins lose 8-2. Odds of a 15-game losing streak 2,348,470,670:1 against[/table] October 7th, 2019: ALDS Game 3 [table] Game State Bottom 2, 0 outs, Twins losing by 1 vs. New York Twins Win Probability 62.6% How'd things look? The only entry on this series where the Twins were favored while losing. Why? The Twins opened the 2nd inning by loading the bases with no outs. This is a scenario where you are highly likely to score multiple runs. What happened? They didn't. Odds of a 16-game losing streak 6,279,333,342:1 against[/table] September 29th, 2020: AL Wild Card Round Game 1 [table] Game State Bottom 5th, 0 outs, Twins ahead by 1 vs. Houston Twins Win Probability 78.4% How'd things look? Twins open the 5th with consecutive walks while already leading. What happened? Strikeout, pop fly, groundout. Twins twitter immediately fears the worst due to the failure to capitalize, and their fears are proven valid. Odds of a 17-game losing streak 29,070,987,697:1 against[/table] September 30th, 2020: AL Wild Card Round Game 2 [table] Game State Bottom 6th, 0 outs, Twins tied vs. Houston Twins Win Probability 57.9% How'd things look? After loading the bases in the first inning and still failing to score, the Twins have done very litte. Still, it's a tie game, and the Twins are coming up to bat as slight favorites. What happened? The bats continued to stay silent, and couldn't overcome a 2-run deficit in the 9th. I cried, and then began writing this article as a coping mechanism. Odds of a 18-game losing streak 69,052,227,309:1 against[/table] Conclusions https://twitter.com/GoTwinkiesGo/status/1311406223019790336

So the computer guys told us this would be smart, we see the trend and we jump on it. Use relievers every day! Let's look at a couple simple stats that are within my grasp. 162 games - average reliever use per game now 3 - put in an opener and it might be 4, but lets not worry about those games where Giminez came in or other extended innings. Just 162 games times three - 486 relief appearances. So we carry 13 pitchers, 5 are starters. 8 relief pitchers divided into 486 means 60 relief appearances per pitcher - forget those who are so valuable that they are out more often. Check out historical use on Baseball Reference - https://www.baseball-reference.com/bullpen/Pitcher#Historical_Usage and you can see the various trends in pitching from the every other game starters of the 1800s to the four man rotation and lots of complete games to five man rotation and a growing requirement for relief pitchers. The more pitchers the less they are used and the more relief we need. Tommy John surgery increases, pitchers are using pitch counts but no one knows whether it is the pitch count or the frequency of pitching that matters. How did Warren Spahn and Juan Marichal and all those great pitchers survive? Must be some stat there that can answer. Trends have always made us smarter. Remember when Coca Cola came out with New Coke and soon the non-stat drinkers rejected it as well as Crystal Pepsi and its clear cola. Or maybe your family had an Edsel - talk about a car of the future! Or more recently we had Netflix divide into Netflix and Quikster - don't remember - understandable. The streaming only branch was flushed down the stream. There was the Apple Newton that was an instant success followed by an instant failure, it just did not deliver, kind of like the sixth man in your bullpen rotation. Then there is that weight reducing fat - Olestra - that Lays used to produce Wow chips! In one year the FDA called a halt, of course the customers did too when they learned that the way that they lost weight was because it induced diarrhea. Kind of like changing pitchers 3 times in an inning induces a coma. Two giants sat down with their marketing stat heads and combined to produce a soda bottle product called Mazagran - coffee tasting soda - within the year the stats called sales numbers forced this Starbuck/Pepsi product off the market. And it would have been an excellent opener to start your day. I will not even comment on the attempt by Colgate - the toothpaste company - to put out a line of frozen foods. Did they clean your teeth when you were done? We will never know. But more recent and perhaps more important to this audience - Playboy decided to drop nudity - where are those geniuses. Did they really believe people bought it for the articles? Well they don't now - those geniuses are back in the minor leagues and nudity is on the rise again. So now we have a trend that created a trend - fifth starters were not much better - if at all - than the bullpen guys so suddenly we evolved to bullpenning. The term does not mean anything, but it is a trend, just like launch angle and increased strikeouts. Does that mean anything to the game? Well strikeout require more pitches which means the pitch count is reached earlier so we must pull the starter and bring in the reliever. More pitches, more game delays, more time before the game ends, longer games and the commissioner wants to figure out how to change this. Good luck. Check out various trends with this excellent set of graphs - https://michaelbein.com/baseball.html then look at the graph on this site for length of games and runs scored - https://www.beyondtheboxscore.com/2015/1/29/7921283/baseball-game-length-visual-analysis and then we ask the question - do people want longer games with less runs scored? Do people want to see more pitchers and less runners on base? As an old guy I love Mike Trout - “The two biggest stats to me are runs scored and RBI,’’ says two-time MVP Mike Trout of the Los Angeles Angels, the game’s greatest player in the midst of his finest season. “I mean, that’s how you win games right, scoring the most runs?’’ Bob Nightengale has an interesting article - https://www.usatoday.com/story/sports/mlb/columnist/bob-nightengale/2018/06/20/mlb-bad-baseball-attendance-strikeouts/718162002/ - that looks at trends and solutions. So, if the trends are hurting baseball, baseball might want to buck the trends, limit relief pitcher use, reduce the innings, move back the fences, reduce the innings. I do not know the answer, but as a former tax accountant I can tell you that numbers can prove many things, but they cannot make the game more enjoyable, unless you are just into APBA, Rotisserie, Fantasy, X-box, etc; nor can they change the human body. Use stats, but don't go too far I really want to see a baseball game - not relays from the bullpen.

Perhaps no Twin better typified the mid-2000s teams than Nick Punto. The nibbliest of the piranhas, Punto played every defensive position over the course of his career except pitcher and catcher, and played most of them better than the average major leaguer. In fact, he had nearly 4000 chances to make a defensive play, and made just 84 errors. Defensive stats have evolved substantially since Punto came into the league, but they’re all fairly unanimous in showing that Punto was a positive asset defensively no matter where he played. The fact that his inclusion in the lineup on a game in and game out basis was as controversial as it was is a great testament to the fact that he was 1) versatile 2) a strong defender and 3) a virtual waste of a plate appearance. For his career, he was about 23 percent worse than the average major league hitter and that includes his inexplicable, galling 2011 when he was 25 above average for the Cardinals after having been 32 percent below average for the 2010 Twins. Perhaps no season serves as a better example of the Punto paradox than 2007, when he was the worst qualified hitter in baseball, but still managed to eek out a positive WAR thanks to his defense and adequate baserunning. Punto’s glove was too good to leave on the bench, the Twins believed, but putting him in the lineup meant sacrificing elsewhere, which ought to sound very similar to the situation the team is facing this year with Miguel Sano. No one is unclear why the Twins want Sano’s bat in the lineup, not after what he showed in his 335 PAs last year. By wRC+, Sano was one of the 10 best hitters in baseball (min 300 PAs) last season, and if that doesn’t buy someone a guaranteed spot in the order, absolutely nothing will. But the presence of Joe Mauer and Trevor Plouffe, and the acquisition of Byung-Ho Park means that Sano will now be judged by both his offense and his performance in the outfield. The Twins may have hoped Sano would be further along in his development as an outfielder by this point, but there was no way he was going to be anything other than a work in progress for most of 2016. His ill-conceived dive on Tuesday night that cost the Twins a run showed that his instincts are still coming along, but he’s already gotten on base multiple times in one game twice in the three games so far this season, so the yin and yang of 2016 Miguel Sano is already on full display. Much as we wondered how bad Punto’s offense could be before Ron Gardenhire would stop penciling him in the lineup, the question that will almost certainly face Paul Molitor at points his season is how bad can Sano be in the outfield while still providing enough of a reason to keep him in the lineup. In 2009, Adam Dunn turned in the worst defensive season by any outfielder since 2000. He was 44 runs below replacement defensively that year, though he split time between the outfield corners and first base, where he was also execrable. He hit 38 home runs, walked in over 17 percent of his plate appearances, and was 42 percent better than league average on offense to compensate for being an unhidable butcher in the field, and managed to produce a 1.1 WAR that season. Clearly the Nationals were hoping for an overall better result from Dunn in his first year with the team, but it’s hard to argue that they got anything other than what they should have expected. If Sano matches Dunn, he’ll still be an offensive star, but he’ll give the Twins less overall value in 162 games than he did 80 last year. Is that acceptable? It’s certainly not desirable, but will the cumulative effect of having Park, Mauer, and Plouffe in the order along with Sano produce the surplus value the Twins want? Possible, but still suboptimal even still. There is a pretty clear model for the player the Twins would like Sano to be as long as he’s learning the outfield: Manny Ramirez. Ramirez wasn’t just bad when he was learning his position, he was hilariously terrible in the field for most of his career, and yet, since 2000, Ramirez is one of only two players to have a season where he was worth -25 runs or worse defensively and still post a WAR of 2.9 or higher. He did four times (Hideki Matsui was the only other to do it, and he did it just once) between 2000 and when his career functionally ended in 2009. Ramirez’s 2005 season was the sixth worst defensive player-season of the new millennium at -32.6 runs below replacement, but he hit 45 home runs, was 52 percent above league average offensively, and helped anchor a Red Sox offense that scored an MLB-best 910 runs. 2.9 WAR certainly wasn’t his high water mark, but it was good enough to help the Sox secure a playoff spot. Unlike Dunn -- whose offensive profile more closely matches Sano’s than Ramirez’s does -- Ramirez wasn’t a strictly three true outcomes threat that season, as he hit .292/.388/.594 to help drive up his overall value. If Sano ends up being the next Manny Ramirez, the Twins should be elated even with the accompanying defensive frailties, but betting on that career arc is awfully optimistic. As mentioned above, Sano’s skill set is similar to Dunn’s: Hit for great power, walk a lot based on the fear you instill in opposing pitchers, and strike out an impressively high number of times, which means that in order to produce the type of value the Twins need Sano to produce to be competitive this year -- and, in truth, in the future as well -- he’ll either need to keep his defensive value above -20 runs below replacement or add a high batting average to his offensive arsenal. It wouldn’t surprise me a bit for this to be the worst season of Sano’s career. He ought to get better and better in the outfield as he gets a feel for different parks and as his instincts kick in, which means that even if his offense stagnates (if you can call repeated seasons at 40 percent above average stagnation) his overall value will continue to rise. Living between 10 and 20 runs below replacement would position him in the Ryan Bruan or Giancarlo Stanton realm of being far better on offense than on defense, but valuable enough in total to make a serious MVP case in years of exemplary offensive performance.

I tend to spend my mornings with four things: a bowl of cereal, a cup of tea at hand, my MLB At Bat app and a folded over copy of The Economist magazine. As I gorge on both carbs and information, I'm happy to know that the Twins have room to grow and that seemingly every night someone out does themselves (even if someone else does worse). But I'm also happy to balance the Twins in one corner of my mind and world events (like Greek debt defaults and Indian corruption trials) are squarely in another. So it's a little startling when those two worlds actually do collide. At the start of this month The Economist published an article entitled "Every step they take" about the ease with which advanced statistics have permeated the sports world, in particular the plethora of statistics that the MLB Network can show on an average broadcast. From pitch speed and curvature to fielder reactions, speed and routes. None of what they covered will be terribly groundbreaking to the most hardcore of Twins fans. The folks who analyze the team and read amateur blogs know about such statistics already, there's little that I (or a bunch of International foreign affairs nerds) can add to it. Not the cover of this article, but appropriate. But what I can say is that when the world of statistical analysis collides with the world of my breakfast table, we've crossed a threshold. This isn't references to Moneyball or "advanced statistics" that make oblique allusions to the wider world of baseball knowledge. This is detail. This is substance. This is specific and detailed analysis smack dab in the middle of mainstream media. The Twins may be using more advanced metrics under Paul Molitor, they may not. Terry Ryan may have turned a corner in his evaluation systems, he may not. But when statistical analysis is offered openly to students in Sri Lanka and professors in Peru and bureaucrats in Burundi and shopkeepers in Slovakia, then you are past the "wait and see" stage. You're past the "consider all aspects" phase. Statistical analysis isn't advanced knowledge, it's mainstream, and if evaluations and judgement aren't made on those metrics...well... Put it this way, I'd like the front office of my favorite baseball team to be as well informed about recent advances in baseball statistics as people half-way around the world who've never seen a game, but who love to read. I hope they are, I hope Ryan and Molitor talk about such topics, I hope I'm underestimating them. And if not...they can always come over to my house for breakfast.

Anyone who seeks out more writing about the Minnesota Twins, clearly cares about the team. They have opinions aplenty about the best direction the franchise could take. They think about it, they weigh pros and cons, and they argue with passion when they feel like they are right. In that regard, there's very little that separates blog readers from the Twins front office. But in the last few weeks a font of frustration has welled up, particularly as regards recent roster decisions. I am no kind of astute baseball analyst (I mean, a large number of my posts turn in to abstract satires of North Korea...), but I think I know why this is. It all comes back to a key division between baseball fans: the fans of words, and the fans of numbers. Fans of words like the story telling aspect of the game: the heartwarming narrative of a player coming into his own or coming back from injury; the mythical prowess of a 100 mile per hour pitcher or a Ruthian Home Run machine; the emotional love of the game. Fans of numbers like the statistical and factual aspect of the game: the value a player brings to the field, their role in creating runs and wins, their failure to avoid defeats, the logical appreciation of the game and its players. While I normally think about the separation between fans within the stands, the same split occurs when we try to evaluate players, and can be expanded to apply to when anyone evaluates someone else. Think of it like this: if you work in a job where you get performance reviews (and I'm struggling to think of a job where you wouldn't), your boss might highlight your productivity by saying something like this: Or they could highlight it by saying something like this: But in reality, they probably have a mix of both the words that colleagues use to describe you and the statistics that they can measure. (As a school teacher I admittedly have no earthly clue what business meetings sound like, but I do know that I'd rather be judged by both comments from other teachers and student performance on standardized tests rather than just one. I suppose I'm hoping that other people have similarly rational evaluations.) http://www.gstatic.com/tv/thumb/dvdboxart/30400/p30400_d_v7_aa.jpgThat's really what we argue about when we talk about who is ready and who isn't ready for the major leagues. We're used to the Twins scouting department (a more word savvy crew) running the show, basing judgements off of what they see in the minors and what the manager sees during Spring Training. Meanwhile, many of the fans (including those who seek out articles to read on-line) are hungry for a more number-friendly crew. But for as much as we talk about the Twins' statistical analyses (or lack thereof) as a catchall for the team's failings, we have to remember that there are benefits and drawbacks to both ways of evaluating people. Word lovers may be able to accurately describe a person's character, demeanor, attitude and potential, but they risk falling so in love with a concept of performance that actual performance means nothing. (After all, if word lovers like me ran teams, somebody would be feeling a nine man team of Air Buds) Number lovers may have a more accurate measurement of a player's performance on the field, comparisons with others their own age, and insights into areas for growth, but they risk turning an individual strength or weakness into a career defining fact. (After all, if statistical measurements of skills were 100% infallible, Moneyball favorite Jeremy Brown would have been an All-Star, and Ryan Leaf would have proven more mature, intelligent and effective than Dan Marino). http://upload.wikimedia.org/wikipedia/commons/thumb/a/ae/Aristotle_Altemps_Inv8575.jpg/765px-Aristotle_Altemps_Inv8575.jpgThe best case scenario is as old as Aristotle: moderation in all things and extremity in none. Evaluations should mix words and numbers, and while there's certainly anecdotal evidence to suggest the Twins could use more numbers, that doesn't mean that words are totally irrelevant to evaluating a player. There is far more that unites we Twins fans and the team management than divides us. Fans and management want a good team. We may have different ways of approaching that goal, but just as we accept both written and statistical performance reviews in our own jobs, just as we enjoy a beer with fans who talk about VORP as much as those who talk about "intangibles", we are better when we use both together.