In addition to plate appearances versus at-bats, I think one also needs to consider the, what I'll call the X-Factor, for lack of a better word. A rain shortened game means x fewer plate appearances. An injury or ejection or substitution means fewer plate appearances. And one "biggie" that players like you and me don't have control over is last minute lineup changes, where the guy we pick isn't even in the starting lineup because... whatever. And then he comes off the bench and gets 1 plate appearance as a pinch hitter. I've seen streaks ended from ALL of these examples. My 15 game one ended when Freddie Freeman didn't get a hit but was taken out in 6th or 7th because of a lingering injury (for which he is now on the DL I've heard). I mean it might have ended anyway, but who knows with one or maybe two more chances. I saw the current leader at the time have his/her streak ended when /she picked Escobar of the Washington Nationals who struck out in the top of the first inning, and pulled something in his oblique, and boom, streak over.

I just think the 80% number is a best scenario case, day in and day out, number, and I don't think it's a fair representation of the actual odds over the course of a long streak where things like I just mentioned have a very good shot of occurring.

I mean I don't even know how to do the math on this, but what if "ordinarily" there is a 80% of success every day, but over the course of 57 days there is an 80% chance that 3 of those days will have only a 30% (because the player only gets one plate appearance)??? I assume that what one would do would be figure that into the overall odds and thus lower the "ordinary" 80% every day to something lower, but these are the types of assumptions that are going to be hard to make even if we can then come up with the right math. As I said, for me, it just "feels" more like about a 2 out 3 chance, whereas for you it may feel like a 4 out of 5 chance, and I'm not sure how we could verify which is actually correct. I suspect it's probably somewhere in the middle, but who knows, perhaps both are optimistic given the variables that occur in a baseball game.

I just did some numbers based on the top 30 batters of 2014, and if I did the math right, it does look like your 80% number is closer than my 2/3 number. I came up with about a 75% chance of a hit over the course of a season. So I guess you're close. I'm just still not sure that takes into account the variables I mentioned above though, such as picking a player who doesn't end up in the starting lineup and gets one pinch hit appearance.

But if I did the math right (always a big if) over the entire season of 2014 the elite batters had about a 75% chance of getting a hit in a game they appeared in. That's going to most likely mean the odds are better than winning the lottery, although I didn't run that.

I dunno, I still don't "feel" good about that considering the variables I mentioned. Just feels like something is missing.

EDIT: I chose 30 because there are 30 teams. So my thinking was if one picked the best player on every team, on average, they had about a 75% chance of getting a hit in a game they appeared in based on their cumulative bating average, total games played and total at-bats.

EDIT2: And I ran the odds based on a 75% chance. This really shows how a small change in the per game odds quickly make the odds of a 57 game streak go up. If I did the math right, the odds of getting a 57 streak given a 75% success rate are indeed better than winning the lottery, but they are a whopping 1 in 13 plus million. Quite a big difference from 1 in 300 plus thousand. This just shows how small differences in the per game success rate can make a huge difference in the overall odds of a long streak and why it may be hard to say with certainty that the odds are indeed better than winning the lottery because I don't know that one can say with a lot of certainty that the 80% or 75% or any other number is the "right" number.

It's a good point with respect to the "x-factor" things you mentioned, there definitely are factors which sometimes just can't be planned for (freak injury, last minute lineup changes after the player choices have been locked). I think there may be a model or formula though that does make some estimated guesses at quantifying these things though, like Bryce Harper would have some sort of higher x-factor penalty built into his daily hit probability due to his frequency of injury/ejection. Then, other factors like pitcher matchups and such being equal, you'd maybe more inclined to choosing Dee Gordon on that day instead because he is less likely to have the "x-factor" things happen to him given his injury history and temperament.

I might take some time looking into this and developing a "best fit" formula, this exchange with you has been helpful in thinking about some factors and ways to estimate/quantify them that may change the way I pick players and help optimize things.

Regarding the 80% most days and then 30% on some bullet-dodging x-factor days, we could factor that like this: 0.80^54 * 0.3^3, which would get us to about 1 in 6 million.

The other thing to remember is that in a given season, we probably have at least a couple dozen chances to start streaks and take them to 57, especially with the double-down feature given in. So if we do have a number like 1 in 100 million, over the course of a season with maybe 20-30 chances to start a 57-game streak it would come down to something like 1 in 3 million. Still a long shot, but I like those odds for it being a free game. :)


Yes, I get the same numbers with the 75% daily hit success number, about 1 in 13 million. Small changes in percentages do matter a lot -- but I think that's part of the beauty of the BTS game, people who are really into it can increase their odds and accuracy by taking into account lots of these little things.

I'm not sure if you've seen the movie "Casino" or not, with Robert Deniro -- there's this segment where he's a bookie factoring odds and taking bets and he gets all sorts of inside info about a player -- like what woman he was with the night before that might affect his odds calculations for his bookmaking the next day. I like to fantasize that's me in my completely nerdy computer spreadsheet sabremetric site visitng way (it's much more comfortable than being in the mafia and risking getting my kneecaps bashed. :)

Yeah, I try to optimize my picks too. Similar to your original post in the thread (though I don't follow the road thing because of the X-factor of most players perform better at home may/might mitigate the possible lost AB's due to the bottom of the 9th not being played). For example, I never pick anyone lower than 4 in the order, and only 4 if it's a real good batter/pitcher matchup. I also never pick "against" certain pitchers, no matter how hot the hitter is or his previous success against such a pitcher. I simply will not pick anyone playing against Clayton Kershaw, Felix Hernandez, Max Scherzer, and a few others. And I try and avoid left handed hitters against the better left handed starters (like Sale as an example).

One thing I can't come up with though is how to find injury information on players NOT on the DL. There are a myriad of sources available for players that are on the DL but the Freddie Freeman thing really bothered me because prior to that he had 100% for me (8 for 8) and I just wasn't aware that he had gotten "nicked" in a game prior to that one. Had I known he had a hurt wrist, I never would have picked him that day. Do you know where one can find complete injury information, including the just day to day stuff for which they aren't put on the DL???? (It's true Freddie Freeman was put on the DL, but he wasn't at first. He was just placed there yesterday I think, apparently the wrist injury is worse than first appeared).

But what I do is look at the matchups for the day and then pick out the best 8 or so, then I go into all their individual stats and see who's "hot" lately (both pitchers and hitters) and who's not (for example a hitter who has a .400 average over 30 at bats against a pitcher looks good, but I f I see he's 4 for his last 38, I'll usually shy away from him). But yeah, it is fun, and I have no illusions of winning, I just like playing to beat my own numbers.

15 is personal high now, and I'm going for that. Currently at 6. And BTW, I just looked at my statistics under profile, and I've picked Jose Altuve more than anyone, which surprised me because I haven't picked him in about a month because I finally figured he wasn't having the season he had last year and he had let me down so many times. But second most was Miggy Cabrera, and the funny thing he's only come through 61% of the time for me. LOL, I must really suck to be able to pick Miggy and only get a 61% success rate. That guys hits all the time.

I don't know of any free newsfeed that keeps concise, searchable info other than ESPN sometimes has small blurbs on minor injury statuses -- but they aren't very reliable and not updated frequently or for every player. There might be some hardcore fantasy sports sites that keep tabs on this info, but I have a feeling many of those are pay subscription-based. Personally, I just try to keep up with happenings around the league as best I can by watching with my mlb.tv subscription, picking up little tidbits from announcers, visiting team message boards to see what people are talking about. I'm not doing it solely to pick up BTS info, I just do it cause I love the game of baseball and in keeping up with my interest I pick up a lot of useful info that can influence my BTS decisions. Before locking in my BTS pick though, I often do check the day before's boxscore to make sure the guy was playing and also do a quick google news search on injury statuses for the guy. That sometimes can reveal nagging or small injuries that aren't enough to keep a player out of a game but which might affect his performance or possibilities of subsitition / re-injury.

I started playing BTS since the beginning, was very casual about it at first then started getting more into it as my interest in sabermetrics increased around late 2000's. This season I've been more lax about my picks, but might try my hand at creating that formula checklist I mentioned earlier so I don't spend too much time optimizing and researching every pick. I've had a couple streaks run into the 30's, a handful into the 20's, and every season have one or two that at least go into the teens. It's fun, but even with the best estimate of odds it still requires an inordinate amount of luck to get any significant prizes. Sort of like with anything, really. We can't change the hand we're dealt -- we can only try to make the best of what we've got. In that way, I don't focus too much on length of streak as opposed to thinking about if the picks I've made were reasonable or not, in spite of what the results may have been.

I'm not sure if you've encountered it yet or not, but MLB sometimes runs a BTS in a Day promotion, where you pick 57 players on a single day and if all of them get a hit you win $5.6 million. That's another fun one, but the odds on that one must be pretty astronomical.

So, on Saturday, the holder of the longest streak of the year was at 43 and counting, but he picked Joe Panik on Saturday and it ended the streak at 43. (I can't recall the name but it was something like confoque or something like that with some numbers).

Anyway, if one looks at the leaderboard now, this person and his/her streak has completely disappeared from the leaderboard, and a different leader is showing (with a long streak of 41, although there is a current one of 40 that may take over the lead).

Yet I can't find any announcement on MLB dot com regarding an explanation. So I wanted to ask you if 1)you noticed this and 2) whether you've heard any explanation or rumors as to why this person was removed from the leaderboard. I've never seen this happen before and it just seems that not making an announcement (I know I can't be the only one that noticed) would be worse PR than whatever the explanation is as to what happened because by not announcing it there will probably be rumors and wild speculation. Even if the person had successfully hacked the website and "cheated" to get the streak to 43, it seems to me it would be better PR to announce that MLB dot com and discovered this and removed him. And if s/he was removed because they weren't eligible, it seems like an announcement would be a good way to remind people to check the rules for eligibility.

Not making an announcement at just seems odd to me.

I hadn't noticed that. I'm not really sure if it was an intentional thing or not, perhaps could just be a bug in their site. MLB.com has been having some hiccups over the past week, so it might just be something related to that.

I don't think so. This person was on the leaderboard for quite some time, we saw the name before it got to the very top because it was the current leader, which we always follow (I say we,. my nephew lives with me, and he follows the BTS game with me and he had an account and was picking daily too but he gave up on picking LOL).

No, something happened. This person had a streak, was on the leaderboard all the way until it got to 43, and then the streak snapped (Joe Panik got no hit in Saturday's game) and then.... poof.... disappeared from the leaderboard completely.

Here are two screenshots showing the overall leaderboard and the current leaderboard. This person was leading the current leaderboard for awhile before he took over the lead of the overall board and we saw the name there for quite a few days before it disappeared. We always check the current leaderboard to see who is getting close (and to root against them LOL cause, you know, I'm a selfish bastard).

ADDED IN EDIT: And it brought up a point I was wondering about ever since I read this little blurb because I wasn't sure why you were seeing something different from what I was seeing and was wondering if it was because you're in Japan. But as you can see on the overall leader board, my 15 game streak ranks somewhere in the 20,000's. And my current 2 game streak is ranked at 100K+, which is what I always see until the streak reaches about 5 or so. I think there are more people playing than you may have been thinking, but maybe it's only showing up with 37K even when you're at 0 because it's in Japan???? I dunno.

ADDED 2: We were thinking it might be because at some point I know MLB contacts the person with the highest streak (because the highest streak for the year wins $10,000 if no one gets to 57) to verify eligibility and perhaps the person wasn't eligible. But if that were the case it seems like an excellent opportunity for MLB to make an announcement urging players to check eligibility because I simply can't imagine how frustrated I might be if I poured over this thing long enough to build up a long streak only to discover I was ineligible. But because of the way the site was acting (as you said you noticed it too) we thought too it might have been hacking attempts and this person was "caught". It's got us both very curious.

NOTE: For any readers following this conversation, what I mean by eligibility is there are rules regarding who can play/win and who can't. I'll post them here for info purposes:

Yeah, it's a bit weird with the 37,000th showing when I had a zero streak. I currently have a 3-game streak and it shows me at like 44,000+. Perhaps it's possible that a ton of people had their streaks reset to zero at the same time? Another thing is that I'm usually checking leaderboards from my iPhone app -- maybe the servers aren't connected or there's some weirdness with how they're displaying?

I don't really know, MLB.com has always been a little iffy with their site stability and programming. I have subscribed to MLB.tv for about a decade and they still have technical difficulties at the start of every season. I've kind of gotten used to the fact that the stuff they put up isn't always rock-steady and reliable with how it should function.

The eligibility thing may be a possibility -- I'm sure most people don't read and digest the fine print before entering into contests like this. Even if the previous leader is a US resident and 18 or over in age, it's still possible he may have a direct relative who works for some MLB-affiliate or something which would disqualify them from eligibility.

In case you are wondering about my eligibility, I am an American expat that primarily lives in Japan but I maintain US citizenship and residence.

Oh gosh no, sorry, I should've made that clear above. No, we have residents all over the world, and especially in the countries of Japan, North Korea, and Germany where we still have significant military presences. I myself was stationed in West Germany for over two years (waaaaay back when there were two Germany's LOL).

No, I was just posting for info purposes, and wondering out loud if maybe that was what caused that person to be removed. I know it's probably none of my business what happened, but man my curiosity is peaked. LOL. I don't think it would have been near as peaked if I hadn't seen what I saw last week on the site, but man last week it was really flaky, and I've never experienced that before there. You see a flaky acting website, and then a leader of the website's contest removed, and your curiosity just goes bananas. LOL.

I'll shut up about it now. LOL. This thread has already gone further than it should have because I keep running my mouth (fingers??). LOL

It's all good! I'm glad this thread is still going, although it does appear that we are a pretty rare breed around these parts and the convo is just ping-ponging between us. :)

Now you've got me thinking about that removal too -- I don't recall ever having seen that happen before in all the years I've been playing the game.