Date: Wed May 7 08:08:16 2008
Sender: Chris Oaks
I want to do a feature in CCEL, and wondered what people thought were the
luckiest aspects of the sim. Be concise, and please please please don't turn
this into your personal bitch and moan session. Mostly I'm looking for the
things we have the least control over, like kick returns, for example, where
you can only choose one setting: who your returners are.
So, I guess I'll start things off with:
- Kick and punt returns for TDs
- Interception and fumble returns for TD (no control at all here)
Date: Wed May 7 09:21:00 2008
Sender: AJ Perko
Hail 'Freakin' Mary & End Zone Pass-
and the defense that defends it.
Date: Wed May 7 09:22:15 2008
Sender: AJ Perko
I think Extra-Points too.
(in Pro) my 10 accuracy kicker misses just as many as my 4 accuracy kicker
Date: Wed May 7 10:31:38 2008
Sender: Loren Smith
To be honest, I think that the odds of a punt return or kickoff return is
partially dependent on the kicker's accuracy attribute.
Date: Wed May 7 12:47:07 2008
Sender: AJ Perko
the returner too... is definately an issue.
I've gotten 4 or 5 Kick return TD's in a seasons when I had a 9rs 9spd guy
returning before
Date: Wed May 7 22:03:25 2008
Sender: Lord Bremen
Returns.
Fumbles.
End of game logic (I consider it luck if it backfires half the time).
Drops (Having 9-10 Ps doesn't seem to affect this much, nor does training con.
Some games it just happens).
(Less so):
Long Passes
Punts (Not sure what makes non-spunts, but they are pretty common).
Date: Thu May 8 10:17:04 2008
Sender: Jamie Davies
End of half logic.
When you are pinned inside your own 10, with less than :20 seconds in the half,
you should not be passing and calling timeouts.
Date: Thu May 8 10:34:02 2008
Sender: Kyle Mayhugh
I think it depends on what how we are defining luck.
Are there things we have no control over? Yes.
Are there things we have some control over, but not as much as coaches would
like? Yes.
But even among the things we have more control over, there is always going to
be a certain amount of randomness. And wherever there is randomness, there will
be human beings trying to find patterns in it and finding some that aren't
there. It's the nature of human perception.
The main thing we have no control over is end-of-game logic. It's a cause of a
lot of frustration, but I don't think it costs as many games as people think.
It mostly robs teams of small chances to comebacks that they probably wouldn't
have converted anyways.
In baseball, I'm a huge fan of a concept called "Win Expectancy," which
essentially measures the value of plays by the change in odds of either team
winning, based on historical data. Fangraphs.com is the site that tracks these
live, and it's amazing. In the college sim, a bad EoG moment usually turns a
15% chance of winning into a 12% chance, for example, but it doesn't really
turn a win into a loss.
For the second category, kickoff returns are often cited as luck. Yes and no.
Teams choose their kick returners and their kickers, and both do make a
difference in the odds of a kick return hitting for a touchdown. If a coach
makes a decision that increases the odds of a return hitting on him from 1% to
2.5% (say, by not spending enough recruiting points on a better kicker) and
that happens to hit twice in one game, is it luck? Yes and no.
The third category is the biggest one. It's a lot like poker. All you can do is
shape the odds in your favor and hope for the best. There are people out
there who believe a lot of things that are provably false based on odds. I
suspect it is the same with the sim. Human perception is notoriously bad at
three things that are key here: 1) Understanding statistical distribution, 2)
Telling the difference between randomness and actual patterns (this is sort of
related to 1) and 3) Keeping accurate records of results.
First, a lot of people don't understand that a 50/50 split is going to go for
or against you on long runs a lot. I always recall some lessons from sociology
classes on the flaws in how humans perceive randomness. Ask a person to put
together a string of random digits, and they will almost never have the same
digit repeating, whereas a true random string frequently have it.
The second point is that we are way too quick to jump to conclusions. If we
have a lot of penalties in two or three games against a certain type of team,
we are too quick to assume there is a pattern.
The third is excitement bias. Unusual or exciting events stick out in our
memory much more than mundane ones and are given too much weight in making
conclusions.
Whenever soemone comes out with a theory about the sim, without an actual,
valid statistical study, take it with a huge grain of salt. We just aren't
equipped to make those conclusions.
Date: Thu May 8 12:13:01 2008
Sender: Chris Oaks
My plan was to be tongue-in-cheek with the PR for those very reasons, Kyle. We
always see losing 51-3 as an enormous string of bad luck ... but no one ever
thinks their 51-3 victory is an enormous string of good luck, which it is,
basically. Sure, there is skill in maximizing your ability to succeed (playing
a 9 PS QB instead of a 2 PS QB, etc.), but when you're talking a random number
generator, there's always luck involved.
Date: Thu May 8 19:28:00 2008
Sender: AJ Perko
"The main thing we have no control over is end-of-game logic. It's a cause of a
lot of frustration, but I don't think it costs as many games as people think.
It mostly robs teams of small chances to comebacks that they probably wouldn't
have converted anyways."
50/50 or not I've seen more than a few playoffs and championship games ended
with END ZONE PASS.
DEL is offensively weighted IMO, there are many games which finish 42-35 etc...
I've seen END ZONE PASS from the 6 yardline out. Teams score at over 50% for
sure when inside your 10. That would be more than a small chance.
Date: Fri May 9 18:54:48 2008
Sender: Andy Dolphin
In terms of the luck factors that affect the most games, that's turnover
margin. In real-life and in DEL, turnover margin is one of the strongest
statistical factors correlated with win-loss ratio, but pretty much luck will
determine if you're +2 or -2 on any one day.
The luck factor most strongly affecting one game would be rare single plays,
such as hail marys, touchdown returns, 80-yard passes, etc.
Date: Fri May 9 19:19:46 2008
Sender: AJ Perko
Andy-
since your here. We can influence turnovers to a degree though right? Whether
by nature of players, or style of defense/offense.
I usually have very few, I don't commit many either. It wasn't always that way
only the past few seasons since I switched some things.
Is that imagination?
Date: Sat May 10 18:14:43 2008
Sender: Andy Dolphin
Absolutely. However, an event that happens only a couple times per game is
more luck-dependent than one that happens dozens (like tackle attempts).
Date: Sat May 10 21:18:21 2008
Sender: Lord Bremen
Andy, while you're here, fumbles are depending upon Rs right? Not Ps as some
people seem to think?
Dolphin Simulation Games is not responsible for the content of posts.
Please report any offensive messages to help@dolphinsim.com.
