A Fun Thought

Dave · April 16, 2009 at 11:45 am · Filed Under Mariners 

Whether the M’s deserve to be 7-2 right now or not, the fact is that those wins don’t get taken off the board. They’re in the bank, and they aren’t going anywhere.

Because of that, even if you haven’t changed your opinion one iota about the strength of the roster (and honestly, you shouldn’t have changed it much – nine games is too small of a sample to mean much), you need to add three wins to whatever you thought the team’s final record was going to be. Math requires you to.

You thought they were a 75 win team on Opening Day? That would be a .463 winning percentage. If they play .463 ball over the rest of the season, they’ll win 71 more games. 71 + 7 = 78.

You though they were a 78 win team on Opening Day (hey, me too!)? That would be a .481 winning percentage. If they play .481 ball over the rest of the season, they’ll win 74 more games. 74 + 7 = 81.

You can do this for basically any expected record. Almost everyone should just add three wins to their expected record to find their new expected record. If you were really high on this team and thought they would win 90, you only add two wins (.555 * 153 = 85 + 7 = 92).

Most of us, we’ll add three wins. So, now, I “expect” the M’s to finish 81-81, based on their current roster, assuming no injuries/trades/etc…

Given that, I’d say it’s likely – not possible, likely – that the team will still be playing meaningful baseball in September. Seriously – get ready for some kind of pennant race. The M’s are in this thing, and barring a summer sell-off of all the expiring contracts, they should be all year.

Comments

67 Responses to “A Fun Thought”

  1. sass on April 16th, 2009 11:49 am

    Not to mention that 5 of those wins are against division rivals. As was discussed last year, those are more important, in that being dominant in those is a good way to win the division, if all other things are more or less equal.

  2. BrianV on April 16th, 2009 11:51 am

    Many thought this team would be fun to watch going into the season, and it’s turned out to be the case. No, they’re not this good (and please, let’s not succumb to “they should be 8-1 because of the blown save!), they’ve gotten lucky, and some guys are playing above they’re heads.

    but man, it’s been a blast watching this outfield, Felix and especially Bedard, and even Griffey, whose signing I was against.

    Stranger things have happened than a ~.500 team playing a bit above its head and taking a weak division, and this looks like it will be a really fun summer.

  3. hub on April 16th, 2009 12:14 pm

    At what point is it feasible for the M’s to become potential ‘buyers’ this summer? And if so, what position is the most likely to be upgraded?

  4. JakeSuds on April 16th, 2009 12:19 pm

    It is great to imagine Safeco packed every night during a playoff run… I’ll be there.

  5. stevie_j13 on April 16th, 2009 12:22 pm

    I’m not sure I entirely understand this point (which probably has more to do with having taken only one stats class six years ago). Regression to the mean recently justified USSMariner’s lack of panic over Silva, and the results were much better in game two. By the same logic, the more the M’s win, shouldn’t I become more and more fearful that a massive, team-wide slump is coming? If a 7-2 team has 75-win talent, isn’t a 4-10 stretch just around the corner?

  6. Noonan on April 16th, 2009 12:23 pm

    I don’t understand. Mostly, my lack of understanding is probably due to my above average lifetime consumption level of alcohol and lead-based paint chips. What I don’t get is this: we can expect someone like Chavez’s batting average to regulate back to it’s norms as the season progresses. Why can’t we expect the same with our winning percentage? Isn’t 9 games of 162 too few samples to adjust the predicted win total for the year?

  7. Noonan on April 16th, 2009 12:25 pm

    stevie_j13 – well put. I wish i had seen your post before hitting submit.

  8. arbeck on April 16th, 2009 12:27 pm

    Noonan and Stevie,

    Future results have no memory of the past. If I flip a fair coin and receive heads 10 times in a row. Then I proceed to flip it 1000 more times. I’m probably going to end up with about 510 heads versus 500 tails.

  9. jld on April 16th, 2009 12:27 pm

    I don’t quite agree that everyone has to tack 3 wins on at this point. If someone thought this was going to be a .500 team, there isn’t necessarily a reason to change that now.

    The M’s could go 2-7 over the next 9 games. At the end of the year won’t matter if a .500 team won every other game or had an 81 game winning streak followed by an 81 game losing streak.

  10. Luc on April 16th, 2009 12:29 pm

    Streaks this long are rare, whether it’s a losing streak or a winning streak. I think it’s reasonable to think that streaking for six in a row at the beginning of the season is more than enough to adjust your predictions at +3 wins. Makes sense to me.

  11. Utah911 on April 16th, 2009 12:33 pm

    If the Mariners can play .600 ball throughout the injuries of Oakland’s starting rotation and Angel’s starting rotation, there’s going to be some serious competition in late August and Sept. Clutch hitting and injuries will be key, but I am starting to get confidence!

    Silva giving up 2 runs?? are you kidding me??

  12. Noonan on April 16th, 2009 12:34 pm

    arbeck – makes sense, but why isn’t this same reasoning applied to the players averages? Seems like we expect players various averages to regulate out to their expected level over the course of a season.

  13. Gregor on April 16th, 2009 12:37 pm

    If someone thought this was going to be a .500 team, there isn’t necessarily a reason to change that now.

    That’s precisely the point of this post though. If you assume that this is a .500 team, you would expect them to win half their remaining games, which would mean they stay about 5 games above .500 for the rest of the year.

  14. arbeck on April 16th, 2009 12:40 pm

    Noonan,

    It is applied to players averages. Lets say Ichiro gets off to a horrid start and bats .120 over his first 100 at bats. That would give him 12 hits over his first 100 at bats. But Ichiro is really a true talent .330 hitter and will get 550 more at bats over the season. If he regresses to the mean, we would then expect him to get 182 hits in those at bats (.33 x 550). His average for the year would then be .298. Regression to the mean does not mean that he will end with a .330 average.

  15. Viper on April 16th, 2009 12:42 pm

    The problem is a flawed understanding of regression to the mean…

    If a fair coin is flipped heads 10 times in a row, the coin is not more likely to land tails on the 11th try. The probability is still 50-50 on where it will land. Regression to the mean does not mean a hot streak is likely to be followed by a cool streak, but rather that a player or team will be more likely to return to their true talent level.

    This means that if you thought that the team’s true talent level was a .463 winning percentage, you would still expect them to win 46.3% of their remaining games. Since 9 games are already out of the way and have no bearing on future performance, you would expect the team to win 46.3% of the remaining 153 games.

  16. tkight on April 16th, 2009 12:43 pm

    Meaningful September baseball seems like a breath of fresh air. I think a lot of it will center around our much improved defense. Like this guy.

  17. Doc Baseball on April 16th, 2009 12:43 pm

    What Dave, and others, are saying is correct statistically speaking. If you thought the M’s were a .500 team, and you had said BEFORE the first game that the M’s might go 7-2 at one point and then 2-7 later, that would be correct and this streak would not influence your expectation. However, once they have gone 7-2, if you want to now estimate the future, that record is already in the past. Estimating the future now — statistically speaking — cannot take that record into account. Unless you think these 9 games have shown the M’s are a different team than you thought, your winning % number should stay the same, but be applied to the remaining games, as Dave did, and as arbeck illustrated. It would only be predictions BEFORE the fact that would place this into an expected range.

  18. domovoi on April 16th, 2009 12:44 pm

    Some of you have fallen victim to what is known as the Gambler’s fallacy.

  19. Steve T on April 16th, 2009 12:46 pm

    Well, you losers can adjust all you want, but *I* thought from the beginning that this was going to be a 126-win team, so I don’t have to adjust nuthin’!

  20. Gregor on April 16th, 2009 12:49 pm

    Some of you have fallen victim to what is known as the Gambler’s fallacy.

    Aka the broadcaster’s fallacy, as in “he’s due for a big hit”.

  21. Ralph_Malph on April 16th, 2009 12:53 pm

    Here’s my example: I flip a coin 30 times a row and it comes up heads every time.

    Noonan or Stevie (no offense)would bet on tails, because that coin is really, really due.

    I would bet on heads, because I think it might be a two headed coin.

  22. domovoi on April 16th, 2009 12:58 pm

    Seems like we expect players various averages to regulate out to their expected level over the course of a season.

    “regression to the mean” only means that we expect the player’s numbers to gradually get closer to what we initially expected. The Law of Large Numbers says if we gave them enough “trials” (e.g., for batters, ABs), then their numbers will end up being very close to their true talent, no matter what they initially started at. The issue is whether or not there are enough “trials.”

    So for example, let’s say hitter A will get 500 at-bats and is a true .300 hitter. If he goes 7/10 in his first 10 at-bats, we’d expect his final average to be (.300*490 + 7)/500 = .308, very close to our prior expectation. If the same hitter goes 70/100 in his first 100 at-bats, even in the unlikely situation where our expectations of him don’t change (i.e. he’s still a true .300), he’ll end up hitting (.3*400 + 70)/500 = .380, which is significantly better than .300, but a lot closer to .300 than the .700 average he displayed in his first 100 at-bats. Give him a couple thousand more at-bats, and he’ll end up at .300.

    So whenever Dave or anyone else says “regression to the mean” they don’t mean the player’s final stats will actually end up being their “mean” (i.e. their true talent level), but rather their final numbers will be closer to their “mean” than they currently are. How close depends on how many games remain.

  23. Noonan on April 16th, 2009 1:00 pm

    but…but… i have a feeling!

  24. Coolalvin206 on April 16th, 2009 1:00 pm

    Hey, whatever happens..its been a fun ride so far. Who would think that on April 16th, the M’s would have the best record in the AL.

  25. eponymous coward on April 16th, 2009 1:02 pm

    Most of us, we’ll add three wins. So, now, I “expect” the M’s to finish 81-81, based on their current roster, assuming no injuries/trades/etc.

    Well, the thing is, I did it more as a range. I expected a 75-85 win team (coming out just under .500). Now, I’d adjust it to being more likely slightly OVER .500, more like 77-87 wins (not adding all 3 wins because a 7-2 start was a scenario partially baked into my initial estimate).

    Just looking at it, it seems the starting pitching and defense will be strengths of the team, assuming that The Interview keeps performing at a high level, King Felix stays on his throne and the rotation does OK. Washburn benefiting from the improved defense in a contract year is somewhat less than surprising, and while our 3/4/5 are pretty generic back end starters, they also aren’t really much better or worse than our 6/7/8 (Olson, Vargas, Jakubauskas).

    The problems that I suspect will have to be addressed if this team stays in contention are as follows, just by looking at the team…

    a) the offense is going to have 2-3 week stretches where not enough people are hitting (Ichiro will have his usual slice of .270/.320/.350 at some point, Endy isn’t going to hit like this all year, Beltre and Branyan will have times where they are swinging at pitches in different time zones, and so on). It’s also really heavily RHB. Figuring out a way to get another LH bat in this lineup in such a way that it doesn’t kill the defense would be nice (in this scenario, Endy becomes a Mark McLemore/Stan Javier supersub in CF/LF).

    b) the bullpen is a mostly RHP who throw plus fastballs with poor command (with the possible exception of Shawn Kelley- I’d like to see him do this for longer, though, before anointing a bullpen ace). I think they’re going to cough up more games like Game 2 before we’re done.

    The nice thing is adding a LH bat and bullpen depth are things that can usually be done pretty easily during the season without mortgaging your farm system (unless you’re Bill Bavasi). As such, I like where this team is. If we’re in good shape by Memorial Day, there’s no reason why we shouldn’t be able to make a run for a playoff spot in a fairly weak division, and a team with a very strong 1-2 punch in the rotation has the capability of going deep in the playoffs (once you have the ability to be picky with your backend starters come playoff time).

  26. domovoi on April 16th, 2009 1:06 pm

    Here’s my example: I flip a coin 30 times a row and it comes up heads every time.

    Noonan or Stevie (no offense)would bet on tails, because that coin is really, really due.

    I would bet on heads, because I think it might be a two headed coin.

    This is a slightly different scenario than what we’ve discussing, as your prior expectations have adjusted. So far we’ve been assuming that our priors haven’t adjusted, specifically, going 7-2 hasn’t changed our initial expectations of how the team would perform. Analogized to your scenario, the 30 heads hasn’t changed our expectation that the coin is still a fair coin.

    Of course, 30 heads probably would make us question whether the coin is fair, so it’s a better analogy of what we should if the Mariners go 23-7. At that point, we probably would adjust our expectations upwards.

  27. Zero Gravitas on April 16th, 2009 1:17 pm

    This is a great point and I also like the contrast to last year, when they started so poorly but kept making the ‘well, it’s still early’ excuse. This team is already 5-0 against the AL West! It’s certainly not too early to talk about the playoff implications of this fast start, given the weakness of the division.

  28. JH on April 16th, 2009 1:18 pm

    Noonan: maybe this will help. Most people say regress TOWARDS the mean, with the understanding that a statistically insignificant but still practically insignificant blip doesn’t have any effect on the probable outcomes of future events. So, when Brian Roberts hit 5 HRs in his first month way back when, everyone knew that he didn’t really have 30HR power, but everyone also knew it would likely be the best power output of his career.

    It’s the same with a player like Endy Chavez. His personal stats aren’t enough to be a huge factor in his overall season line, but it’s enough that if he hits like he’s expected to the rest of the way, his final season line will be a little better than we’d normally expect from him.

  29. Ralph_Malph on April 16th, 2009 1:20 pm

    By the way, coolstandings.com puts the M’s expected record at 81-81 (28.0% chance of winning the division, 3.5% chance of the wild card).

    I’m not sure of their exact methodology, but it’s interesting to see how this has increased with each early win (2 days ago they had us at 77-85/20.5%/3.1%). You can pull up their projections as of each day of the season. As of opening day, they projected the M’s at 68-94.

  30. JH on April 16th, 2009 1:22 pm

    Wow, that was poorly edited. I was going for the difference between regression TOWARDS the mean and regression TO the mean. If the Mariners’ were supposed to be a .450 team, a 7-2 run might not be statistically significant enough to change that expectation of “true probability.” BUT, it’s significant enough to alter the final result.

  31. Soonerman22 on April 16th, 2009 1:24 pm

    Here is another wild thought only 2 starts in.

    Lets say the Mariners are competitive enough this season and they don’t trade Bedard, would you want him re-signed? If Erik Bedard ends up being, in the words of Dennis Green “who we thought he was,” and he was open to coming back would you want him?

    It doesn’t look like Morrow will ever be a starter, and who knows what will happen with RRS. Would you be interested in a rotation with Felix and Erik at the top for the next 4 years, or would you prefer to wash your hands of Bedard and move on with the rest of our lives?

    Just another fun thought/debate.

  32. DizzleChizzle on April 16th, 2009 1:32 pm

    Dave,

    Is it fair to say that winning the AL West is the only chance that the M’s have of making the postseason? How many wins do you forecast it would take to win the division at this point in the season? I know it’s early but I don’t think the Angels are gonna win 100 again.

  33. wabbles on April 16th, 2009 1:38 pm

    So, 85 wins then. Kewl with a capital K. Given what we’ve seen of the Angels juggernaut (Does their outfield REALLY average 34 years old?), September could be very interesting indeed. I guess this is where the FO makes the big bucks. Do we continue building towards the future regardless of our September fortunes, sell off something (one of the expiring contracts) to try making a run for it, sell off something to build more for the future or…Stand Pat?

  34. mln on April 16th, 2009 1:59 pm

    “Because of that, even if you haven’t changed your opinion one iota about the strength of the roster (and honestly, you shouldn’t have changed it much – nine games is too small of a sample to mean much), you need to add three wins to whatever you thought the team’s final record was going to be. Math requires you to.”

    Great. I’ll pencil in the M’s for 119 wins this year then! Woo-hoo. World Series here we come!

  35. eponymous coward on April 16th, 2009 2:01 pm

    Do we continue building towards the future regardless of our September fortunes, sell off something (one of the expiring contracts) to try making a run for it, sell off something to build more for the future or…Stand Pat?

    Um, why would you trade Bedard or Beltre to “make a run for it”?

    Realistically, let’s say we wanted a serviceable OF at the deadline. Look at what Griffey or Winn cost (yes, I know about Griffey in the OF- but the point is he didn’t cost a ton). For a relief pitcher, look at what Ron Villone or Armando Benitez cost. For a back-end rotation starter, check out what it cost for Jamie Moyer (just some examples).

    Really, given that Zdurencik is pretty good at trades from what we can tell (and a marked improvement from the previous administration), I’m not particularly concerned that he’ll give away the farm come July if we have a reasonable shot at contention and an obvious roster hole (and given that Clement’s still in Tacoma, we even have internal options if we need an LH bat). If we give away talent in a trade, I’m pretty confident we’ll get talent back (and it will be a “trade that helps both teams” trade, not a “Bill Bavasi trades away the farm for the wrong player” trade).

  36. Pete Livengood on April 16th, 2009 2:04 pm

    I think they have to consider siging Bedard if they’re at all in contention.

    I was very much against the Bedard trade at the time, unless the M’s knew something I didn’t – that they would be able to re-sign him. It was a horrible trade unless he could be locked up. Given what they gave up, that is still very much the case. Why wouldn’t you make a run at signing him before even considering trading him? And if you’re in contention, why would you trade him (and run up the white flag at the end of July) even if you can’t sign him?

  37. Alaskan on April 16th, 2009 2:13 pm

    Here’s what I don’t like about this statistical concept, as it applies to baseball:
    We’re not talking about a typical coin flip. Rather, let’s imagine we have 29 coins (one for every MLB team besides us), all weighted according to our direct team-to-team comparison. So, our Red Sox coin will come up ‘WIN’ 30% of the time, and our Royals coin will come up ‘WIN’ 70% of the time. If we look at the probability that way, team-by-team rather than simply over 162 games, than the specific teams we played becomes significant.

    For instance, let’s say we are 7-3 after 10 games. Even if we expect this team to be .500 over the entire season, starting out at .700 would only require us to adjust our expectation if we thought the talent of our opponents should lead to less than a .700 average. What if we were playing the Royals the first ten games of the season? We expect to win 70% of the time against them, and 10 games into the year we have. That doesn’t mean that I don’t think we’ll end up at .500 once we’ve played all the other, primarily better, teams.

    If before the season started you asked me what the M’s record would be after 10 games, I would not simply guess 5-5, just because I expect them to finish at 81-81. Rather, I would ask you “who are they playing?”

    Now, this question doesn’t really apply to the M’s, in this case: they are outperforming my expectations against better teams. If you asked me before the season, about these 9 games, I probably would have arrived at 4-5, which is 3 games worse than where we are, so I should raise my expectations for the year by 3. But I don’t understand why we shouldn’t look at the strength of schedule first, before determining that we’re 3 games ahead of where we thought we would be.

  38. stevie_j13 on April 16th, 2009 2:13 pm

    I just want to make sure I am understanding the point of those who are saying I have fallen prey to the “gambler’s fallacy”: If I said at the beginning of the season that the team would win 75 games because of my statistical research, and I understand that teams will go through hot streaks and cold spells in reaching those 75 wins. Hypothetically, those “secondary” stats that predict a 75-win team are still in place, only the team has experienced some good luck and has started strong. Now, I should change my prediction? They lose their next eight games, and now they are a 72-win team? They then win eight of their next ten, and they are a 78-win team again.
    This seems to me to be precisely the small sample size fluctuations that are supposed to be avoided. If I predicted at the beginning of the season that the M’s are going to win 75 wins, that means really that I predicted that you could put a 75-win trend line right through the middle of a graph of the M’s win percentage. It’s not so much about saying that I think the M’s are “due” so much as saying that I think my original prediction still holds.

  39. CCW on April 16th, 2009 2:25 pm

    stevie_j13: Google “gambler’s fallacy”. You just described it. Here it is, from wikipedia:

    “The gambler’s fallacy, also known as the Monte Carlo fallacy or the fallacy of the maturity of chances, is the belief that if deviations from expected behaviour are observed in repeated independent trials of some random process then these deviations are likely to be evened out by opposite deviations in the future.”

    On another note, I’ve seen enough positives from Bedard (and no contrasting negatives) to revise my personal projection of the M’s upwards. They went from an 80-win team to an 82-win team, so I think they’re going to win 84-85 games. Yay!

  40. wildmaxd on April 16th, 2009 2:28 pm

    The justification for add 3 wins is flawed for the simple fact the “odds” and “anticipated outcome” are totally different.

    Just because some one thought they were going to finish the season at .463 doesn’t mean that they have a 46.3% chance of winning each game.

  41. fairweatherfan on April 16th, 2009 2:36 pm

    “you need to add three wins to whatever you thought the team’s final record was going to be. Math requires you to.”

    This also assumes that either “opponent difficulty” for the first nine games is roughly the same as for the remaining 153(seems reasonable in this case – not so if first nine games all happened to be against the Nationals), or that the odds of winning remains the same, regardless of opponent.

    Regardless, the prediction of finishing near .500 ball and playing meaningful games in September, is ample reason for more fans to start tuning in again.

  42. stevie_j13 on April 16th, 2009 2:36 pm

    Also from wikipedia: The Law of Large Numbers is important because it “guarantees” stable long-term results for random events. For example, while a casino may lose money in a single spin of the roulette wheel, its earnings will tend towards a predictable percentage over a large number of spins. Any winning streak by a player will eventually be overcome by the parameters of the game. It is important to remember that the LLN only applies (as the name indicates) when a large number of observations are considered. There is no principle that a small number of observations will converge to the expected value or that a streak of one value will immediately be “balanced” by the others. See the Gambler’s Fallacy.

    Is 162 games a large enough sample? That’s not snarky, but a genuine question.

  43. Uncle Ted on April 16th, 2009 2:37 pm

    I don’t think that Stevie is motivated by the typical Gambler’s fallacy. That is, he’s not saying that there are causal processes that even things out in the end. Rather, he wants to know why some random streak alters the expectations of wins because it happens at the beginning of the season, given that there are random streaks as a standard part of baseball. At least I think this is his question.

    So, Stevie, the point is that the random streaks alter the end expectations regardless of where they happen in the season. To use an extreme example to illustrate the point in the original post, Suppose that the team got of to an 0-50 start, Given your line of reasoning, you’d still say that you’d expect them to win 75 games, or whatever, but for them to do that, they’d have to go 75 and 37 over the rest of the season. That’s a 66% win percentage, aproximately. Which would be a real stretch for a team that you think should win at a 46% clip.

  44. georgmi on April 16th, 2009 2:37 pm

    Yes, you should change your prediction. It is specifically not possible to predict luck, and having some good luck does not increase your likelihood of seeing some bad luck to make up for it.

    Looking forward, seeing continued good luck is exactly as likely as seeing some bad luck. As a result, it can’t factor into your predictions.

    In ’01, we had pretty much nothing but good luck for the first 162 games of the season. Except for that awful, awful night in Cleveland. Last year, we had pretty much nothing but bad luck.

  45. IMFinksPa on April 16th, 2009 2:38 pm

    To be fair, Dave’s post is not entirely accurate unless you assume that the competition faced to date is comparable to the overall level of competition that you anticipated that they would face throughout the season. If, on the other hand, your projection of this team’s talent level was that, against these 3 teams in particular, the M’s could expect to go, say, 5-4, then the over performance is only 2 games. Bear in mind, this not a small sample size issue, it is a sampling issue. The M’s have not faced their entire schedule, they have merely faced a small sample of their entire schedule, a schedule that will include 9 game stretches on the road against such teams as Toronto, Boston, and Tampa, where their expected win total based on their actual talent level may only be 2-7 or 1-8. It is not falling for the gambler’s fallacy if you believe they are due to balance out not as a consequence of regression, but as a consequence of the sample.

  46. Uncle Ted on April 16th, 2009 2:39 pm

    and no, 162 games isn’t a large enough sample. the point of the LLN is that as the number of samples approach infinity, you will tend towards the average.

  47. georgmi on April 16th, 2009 2:42 pm

    I think it’s fair to consider games against Oakland and Anaheim as representative of the average level of competition we’re going to face this year; a quarter of our games will be against those two teams specifically.

  48. joser on April 16th, 2009 2:47 pm

    The other key thing to remember is that the M’s don’t have to be one of the best 8 teams in baseball. They don’t have to be one of the best four in the AL. They just have to have the best record in the AL West. Thus already going 5 for 5 (hopefully 6 for 6) against two of those three other teams is huge. Not only do those wins not get taken off the board for the M’s, those losses don’t get taken off the board for the A’s and Angels. To paraphrase what Dave said on Spokane radio yesterday: if the Mariners can open up a (say) 3 game lead on the other AL West teams and hold on to it into May, it’s going to be hard for anybody in this seemingly weak division to catch them by the end of the summer. Though I wouldn’t discount one of Beane’s patented second-half pushes, and it’s always possible the Rangers will find some pitching under a rock somewhere and surprise everybody.

    And, weirdly enough, the M’s are fairly well built for the playoffs: Bedard and Felix, plus Wasburn and the amazing velcro outfield starting the third game, sets up for a pretty favorable short series.

  49. Mike Snow on April 16th, 2009 2:53 pm

    it’s always possible the Rangers will find some pitching under a rock somewhere

    The only thing you’ll find under a rock in Texas in the summer is a scorpion.

  50. Uncle Ted on April 16th, 2009 3:01 pm

    That’s actually a really good point, Joser. I mean, lets be honest, we’re not competing for the wild card, so really all we need to look at is the comparative records of teams in the AL West.

  51. stevie_j13 on April 16th, 2009 3:02 pm

    I think Uncle Ted summarized my thinking pretty well. If he is right about LLN (and I suspect he might be, as Dave and the others are probably right about the stats), then I should have more license to feel good about this team.
    Part of me just worries about the example of the 2007 Mariners. Many who followed that team, including myself, kept saying, “They’re not this good. They’re not this good.” Then they lost 15 of 17, and it hurt even more because they had outperformed expectations so much and were so close to riding that good luck into the playoffs.

  52. Oolon on April 16th, 2009 3:10 pm

    While I might have to add 3 games due to the hot start, I’m tempted to subtract 3 games due to the team’s fascination with “doing the little things”…

    If we start consistently bunting in the 4th inning of scoreless games with our 6th place hitter to move a runner from 2nd to 3rd – it has to have a negative effect on our total wins. My preseason estimate (77) assumed that Wakamatsu would be a capable manager, these small ball tactics have me wondering.

  53. Wilder83 on April 16th, 2009 3:26 pm

    The real question is, how many games do you take away for the Washington Nationals?!

    This season has already been full of excitement. It really does have a 2001 feel to it. I just hope we don’t absolutely bomb to finish out April.

  54. skjes on April 16th, 2009 3:28 pm

    I’m trying to avoid reading too much into the current series against the Angels. On the one hand, yeah, the M’s are winning and are looking good doing it. On the other hand, the Angels have three starters on the DL, plus Nick Adenhart was killed in a senseless DUI accident (with the entire team dealing with the fact that his next start would have been that Tuesday and his impending funeral). These really are games the M’s are supposed to win.

  55. Ron Stevens on April 16th, 2009 3:40 pm

    Given that they have won 7 of nine,and
    with your premise that they are a .481
    team,I calculated that they have a 32%
    chance of finishing the regular season
    with 82 to 89 wins;they’ve got a 50%
    chance of winning 82 or more.

  56. Ron Stevens on April 16th, 2009 3:52 pm

    correction:
    I should have one less win;
    it should be 81 to 88,and 50%
    chance of winning 81 or more.

  57. Mike Snow on April 16th, 2009 3:56 pm

    Ron, I don’t know what you’re using that affects your formatting, but please remember to use the space bar after you type a comma.

  58. droppedrod on April 16th, 2009 4:07 pm

    It’s been some time since I took a stats class, but does Dave’s logic work when the estimated record for the team isn’t based on a true sample? In other words, is a prediction that the team will finish .500 really a “statistic” in the true sense of the term?

    Saying, before the season starts, that the team will finish 81 and 81, regardless of how you get there, isn’t based on a sample of this team’s past performance. Once you have an adequate sample of the team’s performance, then you can apply Dave’s math. If the team goes .500 through the first 80 games, adding 3 wins to the predicted win total after a 6 game streak in August seems to work better.

  59. DLCheeZ on April 16th, 2009 4:15 pm

    But I don’t understand why we shouldn’t look at the strength of schedule first, before determining that we’re 3 games ahead of where we thought we would be.

    So let’s look at strength of schedule. In April, the Mariners face the following:

    The 2008
    AL West Champs
    AL Central Champs
    AL Central Runners Up
    AL East Champs and AL Pennant Winners

    It’s not all A’s and Tigers. Though let’s acknowledge that the Tigers currently are tied for first in their division.

    As it turns out, the teams we’ve beaten so far all have sub .500 records. Why is that? At least part of that is because the 2009 Mariners are handing out beatings like Halloween candy.

    NOBODY thought the M’s would finish April above .500 and somehow, they’re a 5-8 run away from making that a reality. You’d better believe this team is 3 games ahead of where we thought they would be.

  60. Dylan S on April 16th, 2009 4:51 pm

    “This also assumes that either “opponent difficulty” for the first nine games is roughly the same as for the remaining 153…”

    -fairweatherfan

    “To be fair, Dave’s post is not entirely accurate unless you assume that the competition faced to date is comparable to the overall level of competition that you anticipated that they would face throughout the season”

    -IMFinksPa

    No, Dave’s post doesn’t assume that. It assume that the remaining 153 games are as difficult as the whole 162 games (which they are). The post assumes nothing about the first nine games and gives them no consideration going forward. Dave’s post is making the same prediction he made before the season except now assuming that the M’s play a 153 game season and start off with a record of 7-2.

  61. Gregor on April 16th, 2009 5:15 pm

    No, Dave’s post doesn’t assume that. It assume that the remaining 153 games are as difficult as the whole 162 games (which they are).

    Of course, the assumption you state implies that the first nine games are also as difficult as the whole 162 games, and hence as the remaining 152 games.

  62. beckya57 on April 16th, 2009 6:30 pm

    Calm down, Dave. They will have injuries, after all; for starters, Ichiro’s ulcer could recur, and Jr is only one hamstring pull away from a lengthy DL stint. I agree that this is a .500 team, and that’s a huge improvement from last year (plus they’re a whole lot more fun to watch), but playoff talk is WAY premature. Lots of teams start strong and then regress to their true level later. I think Jack Z and Wakamatsu are building for 2010 or 2011, and that’s exactly what they should be doing.

  63. joser on April 16th, 2009 7:24 pm

    Yeah, because if there’s one thing Dave has been guilty of over the past few years, it’s his excessive optimism.

    The folks that run this blog just can’t win. When they state an unpopular opinion and are proven right, they never have anyone who disagreed come back and acknowledge that. When they’re realistic about the limitations of the team, they’re accused of excessive negativity, of being “haters” or somehow wanting the team to fail.

    And now, when they express a little optimism, they’re told to “calm down.”

    Sometimes I wonder why they bother. (And then I remember: they’re fans.)

    (FWIW, I think the brain trust is building for 2010-11 too; but clearly they’re also trying to balance that with winning now and turning the fanbase around — which, for their employers, is just as important. Hence the Griffey signing.)

  64. joser on April 17th, 2009 1:49 am

    An unfun thought: In the opening days of the 2008 season, the Royals went 6-2 and held first place in the AL Central for 11 days, until April 14th. A week later they had fallen to last place, on their way to a 72-90 record.

  65. DMZ on April 17th, 2009 1:53 am

    That’s not going to happen again since they have Bloomquist to keep them focused and motivated.

  66. joser on April 17th, 2009 12:35 pm

    Oh smacks forehead of course! How could I have overlooked that?

  67. joser on April 17th, 2009 12:36 pm

    Wait, the M’s had Bloomquist last year. Are you saying he’s going to focus the Royals on losing 100+ games this year?

Leave a Reply

You must be logged in to post a comment.