Chat: 9/7/15 1:32am
Chat: 9/7/15 1:32am
Shouts
<p>Mo' Dawg ... I'se expeks yousa right. And what were they doing auditioning Marte in the first place, if this performance didn't win him the job?</p>
<p>Sammy ... for sure it's not realistic in the real world. Too many cars to pass with 2 laps to go. But it's nice to take a last few sips of "live" #Baseball ... :- )</p>
<p>Matt ... leadoff triples in the 30's being worth more than a leadoff homer... what an AWESOME stat. Backs up Mike Hargrove / Ron Fairly in their weird perspective that "a double keeps things perking" compared to "a homer ends the rally." Fascinating. Certainly the pitcher has to keep pitching from the stretch and dividing his attention. #Baseball is a thimble from which a flea may drink and from which an elephant cannot be satisfied...</p>
<p>Rickster - Farquhar's love for his power 89 MPH cutter is comparable to Lou Piniella's love for a pitcher who will "challenge." Farquhar does get grounders (outs) with the pitch, and throws more strikes with it by far compared to his FB and hook, but the problemo is that he makes an inordinate amount of mistakes with it (21% HR per fly ball) and has absolutely no sense of danger with it. John Buck went on a postgame show once and told THE PUBLIC about how brainless Farquhar was about throwing cutters outside. Dr. D could be wrong! :- ) but he is married to the position that Farquhar should be using the cutter on 3-1 counts, down in the zone. Since you axed ...</p>
<p>rick, I would go even further: If McClendon is back in '16, then Marte WILL be our Opening Day SS. Would be willing to bet that some sort of Miller/O'Malley/? is the CF combo, as well. Trumbo is back. Montero is up. Morrison is gone. #Wild card guys are Flores, Kivlehan, Choi and Blash. Guti is back, of course. I'm not kidding guys...Smith could be expendable. If he's back, then Smith/Guti in LF, O'Malley/Miller in CF and as Utility, Cruz in RF, Seager, Marte, Cano + Trumbo/Montero = 10 guys. Two C's leaves room for only one more guy.</p>
<p>It's fun being in the hunt numerically, but there are too many teams between #Houston and the Ms. And they all play each other. </p>
<p>Is there any AL team that can stick with the Dodgers, Cardinals, or even Pirates? KC faltering, #Toronto is scary offensively even without Tulo, but beyond Price they can't match up.</p>
<p>I'd say Marte has the SS job heading into spring 2016. How has Brad looked in CF tonight?</p>
<p>In case anyone is wondering - I want LI because the Elo formula is dElo = K * (W% - xW%) where dElo is the change in Elo rating, K is a weighting term that is based on the importance of the match-up, W% is the actual W% of the match-up, and xW% is the expected W% given the Elo ratings of the two players. I intend to use LI (multiplied by some constant...probably 10) as the K term, I have W%, xW% is recursive and based on current Elo ratings...so LI gets me to Elo ratings directly.</p>
<p>I've also succeeded, today, in appending a play-by-play batter winning percentage onto the record...the crucial first step to computing match-up based player Elo ratings. Now comes the task of implementing Leverage Index...a task I expect will be much easier than it was for me in 2005, when I had to code a monte-carlo simulator on my own in C++. MATLAB has a massive user community, including a "file exchange" where users share codes they've written. There are over two dozen different flavors of MCMCs (Markov Chain Monte-Carlo simulators) available...I just need to figure out how to get my data into a format that would be useful for the MCMC and then run an MCMC for every base/out/inning/side/run differential state and boom...win probability is born...from there, LI is a short hop.</p>
<div class="indented">Reply - mickiholley - 9/25/15 2:33pm<br>you are awesome matt!!</div>
<p>I've tested the run expectancy algorithm on several other seasons just to see...I'm getting the same slight edge to man on third over homer more often than not...but it's always very close. I wonder if it is still true in the modern game...we're getting smarter, I think, about positioning the defense, and I'm wondering how many teams are rightly continuing to do their defensive shifts and conceding that run even with 1 out.</p>
<p>Matt, that whole 1934 No outs, runner on third thing is interesting. I'm assuming the "infield in" thing will increase the chance that the next guy gets a hit and that you then have your run AND a guy on 1st base more frequently that a homer + a hit would account for. It is worth remembering, too, that sometimes that guy on third doesn't score at all in the inning so you get zero runs in those occasions (infrequent) which means in the other occasions the runs scored in the inning exceeded the 1.5588 figure. I wouldn't be surprised if you find that to be the case in some other seasons. it would be interesting to know if managers were more likely to play the infield in back in '34 (and that whole general era) than they are today. Comment on this: If a 1934 manager saw his thrower give up a lead-off triple AND then saw him get an out without giving up a run, would he be likely to walk the 3rd batter in the inning to set up a double play? Weird variables are involved here. And, then again, it might be just random noise.</p>
<p>Interesting lineup choice tonight: Vs. the RHP, in a game we "must" win, McClendon goes with Montero and not Morrison at 1B.</p>
<div class="indented">Reply - anonymous - 9/22/15 3:35pm<br>Will see how it pans out</div>
<p>I can also tell you, just by looking at the tabular breakdown of run expectancy added data for the none on, none out situation, that there were 2 walk-off homers to lead off the bottom of the 9th in the record for 1934. I can tell you that because those homers changed the run expectancy from .5466 to 1 exactly, and the only way that can happen is for the at bat to score a run but produce no additional run expectancy. This is fun.</p>
<p>And...I just (very easily) queried the 139 lead-off homers found in this season's PBP file and determined that the run expectancy (if simply totaled up and averaged) following those homers was, in fact, 0.5470...very close to the original run expectancy for a lead-off at bat. MATLAB is awesome, BTW...that was way...way easier to do in MATLAB than in MySQL</p>
<div class="indented">Reply - anonymous - 9/22/15 3:37pm<br>There websites for this</div>
<p>Amusing fact of the day. In 1934, a lead-off triple was worth more than a lead-off homer. For those that don't know...1934 was a very high scoring era, so the starting run expectancy of an inning back then was .5466 runs. A solo HR has a run expectancy added value of 1, since the base/out state, and thus run expectancy, of the inning didn't change other than that a run already scored. But a lead-off triple changes the base/out state from 0 on, 0 out, to runner at third, no outs...which was worth 1.5588 runs. The difference is 1.0123 runs...or slightly more than a homer. You could put that down to random chance, since the new base/out state only happened 238 times in the roughly 52% of the games that retrosheet has logged in its play by play database...but it is also possible that having a man at third increased a team's odds of getting a second baserunner by influencing the team defensive alignment and/or distracting the pitcher. If the odds of getting any other baserunner do not increase after a solo homer...but do increase after a triple - given that the runner at third with none out scored nearly 100% of the time in 1934, it could actually make it better to get the runner at third than to score him immediately.</p>
<div class="indented">Reply - mickiholley - 9/25/15 2:35pm<br>cool fact! </div>
<p>Question for GM Dectecto: Danny Farquhar has been stellar of late, and credits his cutter for the success: "I'm throwing my cutter for strikes. I'm locating it down, and in and out. It's a huge key to my success." - So, do you trust Danny and his cutter-love going into 2016?</p>
<div class="indented">Reply - anonymous - 9/22/15 3:40pm<br>I woukd. It's like cal.. he changed his stance 100 times to see what felt best for him each season, half season sometimes... in baseball coaching goes far, but these players also know what they like. </div><div class="indented">Reply - mickiholley - 9/25/15 2:37pm<br>he may be finding the groove that will take him far...I would ride it out and see what happens!</div>
<p>Before we get #Houston...we have to play one of the best teams in #Baseball (Royals) for three...if we can get 2 out of 3 against them and #Houston loses during that time...the #Houston series could have meaning.</p>
<div class="indented">Reply - anonymous - 9/22/15 3:41pm<br>Pretty amazing probably too late, but he'll of an effort down stretch..and if they make it... who knows what is possible</div><div class="indented">Reply - mickiholley - 9/25/15 2:38pm<br>anything is possible anymore</div>
<p>-6.5 down to #Houston with 14 to play. With the first loss to #Houston, the curtain comes down. But a 3-0 sweep would leave -3.5 with 11 to play ... which is about what the Seahawks came back on the Cardinals with last year. So, Tuesday night's game is "live" for Dr. D. Maybe they'll get him a Wednesday game too? Hey, very shortly we're going to be in a cold, cloudy winter, but tomorrow night we got #Baseball :- )</p>
<div class="indented">Reply - anonymous - 9/22/15 3:42pm<br>Amazing in 2 months went from wishing baseball to be over to waiting for next game to begin. Huge turnaround </div><div class="indented">Reply - mickiholley - 9/25/15 2:40pm<br>the expectation is palpable</div>
<p>Saw Justin Smoak rip a double in the TOR vs. NYY game tonight, in a huge win for them. Ugh.</p>
<p>Bill James has a stat called Cy Young shares - Felix' 2.42 career "shares" are the equivalent of getting 60.5% of the vote in four distinct seasons, or 121% of the vote twice. Get it?</p>
<p>Felix' 2.42 ranks 13th all-time, and this year he'll leapfrog Johan Santana if he gets 31% of the vote. He needs 64% over the course of his career to pass Sandy Koufax, 73% to pass Tom Glavine and make Top Ten Ever. </p>
<p>Felix will have a tougher time running down Clayton Kershaw in the #9 spot at age 27. Slap me silly.</p>
<p>Clemens leads with 7.5 'shares' and Unit is next at 6.5. No Christy Mathewson, since the award was first given out in 1956. You can safely assume that Christy woulda got 94 shares or so.</p>
<p>You HAD #Baseball knowledge, until you came to reide in this Oceanic Flight 815 flash sideways. - jemanji</p>
<p>Doc...I have #Baseball knowledge...you have considerably better #Baseball wisdom.</p>
<p>Sandfrog lives!<br />
Reply - mojician - 9/21/15 10:26amSandfrog has a brand new album called "Tear". I have no doubts that Sandfrog will book Seattle venues for the Tear tour and then not show up for them. </p>
<p><a href="https://www.youtube.com/watch?v=jOUn1GKTfUQ" target="_parent">https://www.youtube....</a></p>
<div class="indented">Reply - mickiholley - 9/25/15 10:37am<br>awesome</div>
<p>Morales' OPS+ is 121 this season, better than Seager's and Cano's. Would like to see an all-#Seattle tank team. Does anybody top Scott Spiezio's -22 here followed by 120 the next year for St. Louis? And leave us not forget Adrian Beltre: 48 homers the year before, 99 OPS+ his five years here, 140 OPS+ since leaving. Still want the #Seattle GM job?</p>
<div class="indented">Reply - Browns8625 - 9/21/15 9:34am<br>Luckiest team in baseball..sarcasm..cough cough</div><div class="indented">Reply - mojician - 9/21/15 10:00am<br>Oh No! Just thinking about the Sandfrog brings back old scars that I thought had healed over. </div>
<p>Further, Morales has always looked good on paper. Further, we were always pulling for him. -Mojician</p>
<p>I'm a little bit upset about Kendrys Morales. Dude must've tanked on purpose for the Mariners. Today he hit 3 hr and a triple to go with his 105 RBIs on the year. Dude was a bum in teal. What gives?</p>
<div class="indented">Reply - Browns8625 - 9/21/15 9:35am<br>Woukd been nice with current situation</div>
<p>At BJOL, they just now discussed this idea of "targeted" #Baseball stats like 20 wins. Hm, lemme go find that ...</p>
<div class="indented">Reply - anonymous - 9/21/15 4:36pm<br>29nwins used to be fairly common..well at least many more per year. It's quite the big deal these days. FELIX SHOUld BE PROUD</div>