tag:blogger.com,1999:blog-38600807.post8592245123223143832..comments2009-07-03T23:22:02.081-04:00Brian Burkenoreply@blogger.comBlogger8125tag:blogger.com,1999:blog-38600807.post-16716380883212448932009-07-03T23:22:02.081-04:002009-07-03T23:22:02.081-04:00No. I should do that. I&#39;d like to put a little more polish on the end-game model before I do that, however. For example, my current model doesn&#39;t do a good job of guessing when a team will go for it on 4th down, so occasionally it will figure a team will kick a FG when down by 6. Sure, it&#39;s obvious when there&#39;s 1 min left, but what about when there&#39;s 3 minutes left? 4 minutes left?Brian Burkehttp://www.blogger.com/profile/12371470711365236987noreply@blogger.comtag:blogger.com,1999:blog-38600807.post-33120982643085915852009-07-03T22:35:29.427-04:002009-07-03T22:35:29.427-04:00Brian- I&#39;ve just been introduced to your site and am blown away by the work you and the other contributors have done.<br /><br />I&#39;d like to know if anyone has put out research on the success rate of two-minute drives based on starting field position, time left in the game, and points trailing. For example, when a team is trailing by four points (they need a td), and have the ball on their own 26 with 1:32 left to play, what is the success rate of that team scoring the go-ahead td? What about when they only need a field goal to win or tie? It seems to me that knowing these success rates would greatly inform coaches during the end game.<br /><br />I know your Win Probability pretty much does this, but has anyone posted a chart specifically devoted to end-of-game drives?mccarronkhttp://www.blogger.com/profile/06391019249309554677noreply@blogger.comtag:blogger.com,1999:blog-38600807.post-73372481424370270872009-03-28T11:47:00.000-04:002009-03-28T11:47:00.000-04:00This is cool Brian, and thanks for the plug.<BR/>Yell if you get around to doing what Anonymous said, I might be able to keep you from having to reinvent the wheel.idhttps://www.google.com/accounts/o8/id?id=AItOawlNUZhPBX5B90LZw_NEme0fExjUMxMJqzonoreply@blogger.comtag:blogger.com,1999:blog-38600807.post-420380890354677852009-03-28T10:29:00.000-04:002009-03-28T10:29:00.000-04:00Yes, but there are big challenges to doing that, especially for football.Brian Burkehttp://www.blogger.com/profile/12371470711365236987noreply@blogger.comtag:blogger.com,1999:blog-38600807.post-37715133210420614782009-03-28T00:15:00.000-04:002009-03-28T00:15:00.000-04:00Do you have any plans to try to incorporate team quality into the in-game probabilities? That is, a team which is favored by 10 at game start and then goes down by 5 has a much different win prob. than a generic matchup of two average teams. Perhaps use the opening Vegas point spread as an indicator of team strength? So your database would look for historical games in which team A was favored by x points, and trailing by y points with z minutes remaining. I am not sure if this is feasible to calculate, but it would be awesome if it was. (For both football and basketball). Thanks.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-38600807.post-53235032432269977962009-03-27T17:11:00.000-04:002009-03-27T17:11:00.000-04:00Thanks Brian!Griffin Capriohttp://www.blogger.com/profile/03769076370263134103noreply@blogger.comtag:blogger.com,1999:blog-38600807.post-43682315646654082622009-03-27T16:20:00.000-04:002009-03-27T16:20:00.000-04:00http://www.maani.us/xml_charts/index.php<BR/><BR/>or you can click on any of my graphs to get to the site. That's because I'm too cheap to upgrade from the freeware version.Brian Burkehttp://www.blogger.com/profile/12371470711365236987noreply@blogger.comtag:blogger.com,1999:blog-38600807.post-24376354745712014182009-03-27T15:24:00.000-04:002009-03-27T15:24:00.000-04:00Hi,<BR/><BR/>What's the web graphing tool?<BR/><BR/>Thanks,<BR/>GriffinGriffin Capriohttp://www.blogger.com/profile/03769076370263134103noreply@blogger.com