Posted by antonsirius on 8/20/2010 7:06:00 PM (view original):

Posted by daalter on 8/20/2010 3:20:00 PM (view original):

Posted by antonsirius on 8/20/2010 11:44:00 AM (view original):
1) Bucknell had a noticeable IQ advantage on you. About two-thirds of his minutes were played at A+/A; your minutes on offense were split evenly between A- and B-, with a sprinkling of A+. You were a little sharper on defense, but it's still a big edge for Bucknell.

2) You ran a depressingly predictable offense. Despite having multiple players with good-to-excellent PER ratings, the only one you let shoot threes was Hammontree, which means you were actually vulnerable to the old -5/DT trick that Bucknell used against you.

Are those reasons enough to overcome the huge talent gap and 'justify' the upset? Maybe, maybe not. But it's certainly not inexplicable.

anton, your explanation started to intrigue me until I looked and saw that St. John's shot 4-17 from 3p against that -5. With a -5 and a huge sp/ath (not to mention overall talent) disadvantage, that defense really shouldn't have worked.

I agree that there are a couple mitigating factors here, but I don't think it's anywhere near enough to justify the result ... particularly since we see a lot of results like this.

I'd guess that every player on that St. John's roster, except Hammontree, had a 3P setting of -1 or -2. And Hammontree was double-teamed.

I'm not trying to open the door to another round of debate on whether the -5/DT combo should work at all, but the only kind of team it should be even remotely effective against would be a team with only one three point threat on the roster -- and that's what St. John's was running voluntarily, even though they had five or six guys who could conceivably have been jacking up threes with some success.

The problem wasn't just the 4-for-17 against a -5, it was that he only had 17 attempts against a -5. And that's on the offensive distribution, not the engine. He could have gone 7-for-17 on threes, and still would have lost.

Posted by pinkeye on 8/20/2010 5:44:00 PM (view original):

Posted by emy1013 on 8/20/2010 4:41:00 PM (view original):
Uhh, yep, that is a WHOLE lot worse than mine.  Sometimes you can just tell these things and I think Seble may really need to go back and look at the foul situations because I can tell you, it's not working right.  And I'm not saying that just because of my one example above.  It's not off EVERY game, but when it fouls up (pun intended), it's really bad.

P.S.  Wonder how long it'll be until my number one fan shows up with some smart comment........

i've refrained lately, to save your kids the beating.

Posted by davis on 8/21/2010 4:23:00 PM (view original):
I think that the effects of defensive positioning are the single weakest part of the game engine.  Considering how easy it is for even the schmoes I play with (and me, for that matter) to hit open three-pointers from the college line, the default three-point percentage against a -5 should be at least 60% or 70% if we are talking about real college players, and even higher if we are talking about guys with high PE ratings.

On the other hand, the number of inexplicable upsets that I see nowadays is WAY down.  Doesn't make it any less frustrating when you are the victim, but there are far fewer of these types of games than there were under the old engine.  I used to lose to a bad AI team at least once a year, to the point that I would expect the worse every morning when I opened up the play-by-play... nowadays when I lose, usually I can at least look at the rosters and homecourt advantage and understand why.

Posted by antonsirius on 8/21/2010 6:55:00 PM (view original):

Posted by girt25 on 8/20/2010 10:29:00 PM:
I agree that this distro/3pt frequency hurt him. But the talent gap was so incredibly vast, I have a hard time swallowing the notion that a -5 that didn't even really accomplish what it was supposed to is an appropriate justification for this result.

I meant to make this point originally and then forgot:

Whenever I see a thread in the forum decrying a horrible, unrealistic upset, my initial reaction is to approach the game as the underdog and see what I would have done. In this case, the game plan I would have used and the game plan Bucknell appeared to actually use ended up being pretty similar - sagging inside, doubling Hammontree, and focusing distribution on the matchups where I had an IQ advantage that might mitigate the talent disadvantage.

In other words, Bucknell did what I would consider the most logical thing to try and create an upset.

I've seen upsets where that wasn't the case, where it really did look like just a nasty spin of the RNG Wheel of Fortune. This wasn't one of those cases. To me, if you're arguing that this result should never have happened, then you're arguing that game planning is irrelevant -- that there was literally nothing the Bucknell coach could have done to give their team a chance to win the game.

To put it another way, if upsets are going to happen - and I think they should happen - I'd rather see them happen in situations like this, where the game plan looks like it may have made a difference, than in games where some sim-coached team was on auto-pilot and still pulled it out.

Posted by davis on 8/21/2010 4:23:00 PM (view original):
I think that the effects of defensive positioning are the single weakest part of the game engine.  Considering how easy it is for even the schmoes I play with (and me, for that matter) to hit open three-pointers from the college line, the default three-point percentage against a -5 should be at least 60% or 70% if we are talking about real college players, and even higher if we are talking about guys with high PE ratings.

On the other hand, the number of inexplicable upsets that I see nowadays is WAY down.  Doesn't make it any less frustrating when you are the victim, but there are far fewer of these types of games than there were under the old engine.  I used to lose to a bad AI team at least once a year, to the point that I would expect the worse every morning when I opened up the play-by-play... nowadays when I lose, usually I can at least look at the rosters and homecourt advantage and understand why.