[KartyFan]

Matt, what source/analysis do you use to determine whether going for it on 4th down is analytically sound, tossup, or riverboat gambling? Not disagreeing, interested to know.

The best public source I know is CFB4TH (cfb4th.com), which is specifically a college-football fourth-down decision model. It evaluates decisions in terms of win probability, incorporating down and distance, field position, score, clock, timeouts, pregame spread, etc. My characterizations are my own shorthand based on trying to apply the kind of framework that website lays out, accounting for the specific kicking situation. Analytical approaches try to assess the probability of converting vs. the probability of making the kick (or effect of the likely outcome of punting or whatever) and the win probability after each possible outcome. The basic question is whether decisions produce a higher expected chance of winning. As analytical as I tend to be, my own bias is along the lines of what Pritchard purports to be, analytics-guided rather than analytics-driven. If I were a coach (I have no interest and don't think I'd be any good), I don't think I'd go strictly by the book.

Got it and thank you for another great writeup. A few thoughts from my seat in Stanford Stadium:

1. Non-student crowd was sparse, less than the  Hawaii game I think and far less than Miami. But the students turned out really well and made enough noise for the stadium not to feel like a picnic. I hope this isn't just a first game bump and that attending games becomes part of campus life again.

2. The new stadium lighting is very cool. TV does not show it well at all (in fact I found it annoying when I watched highlights later). 

3. As the tides turned in our favor, Steve Frost (PA announcer) took the opportunity to mention that it was 1:15am Eastern Time. Just a reminder to the GT players that they might be getting sleepy!

4. Pass protection was superb. When was the last time we even had enough time to take that many deep shots?

5. Warren's deep passes, wow. Can think of one he missed. So great to see explosive plays.

6. When you can't run the ball, one little thing can derail a drive, and we saw a lot of that in the first half. To become a good team, we need to be able to run the ball and stop the run. We seem closer to the latter.

7. The 4th and 10 and 4th and 11 calls/conversions probably won us this game. Gutsy. Any quibbles with other playcalling are dwarfed by those key plays. Good result doesn't equal good decision, but for those inclined to judge based on outcomes, the good far outweighed the bad.

8. This offense is so much more fun to watch than last year's. It doesn't feel like it will take a minor miracle to score a touchdown. Improvements IMO are QB, pass protection, and playcalling.

9. Feels like we are about 0.5 games ahead of preseason expectations so far. 4 wins is achievable and 5 not entirely out of the question. With our schedule, 4-8 could mean progress depending on what it looks like. 5-7 obviously IS progress and would be a statent that we have a chance to go places. 6 wins still very unlikely but not impossible. We'd need to be better than we have shown so far (or very lucky).

I for one am enjoying this season. Every game is a new discovery and measuring stick, and at least on offense it feels like the staff is maximizing talent pretty well. Let's build on this!

[KartyFan]

Matt, what source/analysis do you use to determine whether going for it on 4th down is analytically sound, tossup, or riverboat gambling? Not disagreeing, interested to know.

The best public source I know is CFB4TH (cfb4th.com), which is specifically a college-football fourth-down decision model. It evaluates decisions in terms of win probability, incorporating down and distance, field position, score, clock, timeouts, pregame spread, etc. My characterizations are my own shorthand based on trying to apply the kind of framework that website lays out, accounting for the specific kicking situation. Analytical approaches try to assess the probability of converting vs. the probability of making the kick (or effect of the likely outcome of punting or whatever) and the win probability after each possible outcome. The basic question is whether decisions produce a higher expected chance of winning. As analytical as I tend to be, my own bias is along the lines of what Pritchard purports to be, analytics-guided rather than analytics-driven. If I were a coach (I have no interest and don't think I'd be any good), I don't think I'd go strictly by the book.

Got it and thank you for another great writeup. A few thoughts from my seat in Stanford Stadium:

1. Non-student crowd was sparse, less than the  Hawaii game I think and far less than Miami. But the students turned out really well and made enough noise for the stadium not to feel like a picnic. I hope this isn't just a first game bump and that attending games becomes part of campus life again.

2. The new stadium lighting is very cool. TV does not show it well at all (in fact I found it annoying when I watched highlights later). 

3. As the tides turned in our favor, Steve Frost (PA announcer) took the opportunity to mention that it was 1:15am Eastern Time. Just a reminder to the GT players that they might be getting sleepy!

4. Pass protection was superb. When was the last time we even had enough time to take that many deep shots?

5. Warren's deep passes, wow. Can think of one he missed. So great to see explosive plays.

6. When you can't run the ball, one little thing can derail a drive, and we saw a lot of that in the first half. To become a good team, we need to be able to run the ball and stop the run. We seem closer to the latter.

7. The 4th and 10 and 4th and 11 calls/conversions probably won us this game. Gutsy. Any quibbles with other playcalling are dwarfed by those key plays. Good result doesn't equal good decision, but for those inclined to judge based on outcomes, the good far outweighed the bad.

8. This offense is so much more fun to watch than last year's. It doesn't feel like it will take a minor miracle to score a touchdown. Improvements IMO are QB, pass protection, and playcalling.

9. Feels like we are about 0.5 games ahead of preseason expectations so far. 4 wins is achievable and 5 not entirely out of the question. With our schedule, 4-8 could mean progress depending on what it looks like. 5-7 obviously IS progress and would be a statent that we have a chance to go places. 6 wins still very unlikely but not impossible. We'd need to be better than we have shown so far (or very lucky).

I for one am enjoying this season. Every game is a new discovery and measuring stick, and at least on offense it feels like the staff is maximizing talent pretty well. Let's build on this!