Win Total Predictions

[bigrabbit]

I feel pretty solid at 5, would normally doubt UVA away but they are beyond awful right now. I think we’re more likely to pull out a 6th somehow than slump to 4, but anything is possible. Who knows, UNC has choked against us the last two times, things happen.

 

[Oldgoldandwhite]

Glad we’re not playing Duke.

 

[BuzzStone]

Glad we’re not playing Duke.

Crazy when I am excited were playing clempson instead of duke in football lol

 

[Northeast Stinger]

Other than uga and a couple of other top ranked teams nationally this may be one of those years where anybody can beat anybody on a given Saturday (or Monday or Thursday or Friday). Is it possible that the portal has stirred the pot that much?

Anyway, my bold prediction for 7 wins holds. In the topsy-turvy season that is unfolding this year a Key coached team could do fairly well, since they will give a consistent effort from week to week.

 

[takethepoints]

One thing: after last night I think a lot of the win/loss predictions need readjusting. Duke no longer can be a predicted win. Clemson can no longer be a predicted loss.

 

[Augusta_Jacket]

One thing: after last night I think a lot of the win/loss predictions need readjusting. Duke no longer can be a predicted win. Clemson can no longer be a predicted loss.

We don't (fortunately) play Dook this year...

 

[takethepoints]

We don't (fortunately) play Dook this year...

Oh.

 

[GetYourBuzzOn]

SP+ was updated today. As expected, GT improved- from -5.4 to -4.6

23.7 (75) for Offense, 28.2 (79) Defense, and -.1 (115) for Special Teams

You can then project out Regular Season Win totals by taking the difference in opponents SP+ ratings and factoring in Home Field Advantage to get a projected point spread.

Once I get my projected point spread I then convert to the Money Line by using a calculator powered by a ton of historical data located behind a paywall on the Unabated site. Convert the Money Line to implied probability, then sum up the probability for the projected win totals.

Here is what I have after week 1:
Louisville (neutral site) / Loss / 0%
SC State / -28.5 / 97.53%
@ Ole Miss / +24 / 4.64%
@ Wake / +11.5 / 19.23%
Bowling Green / -16 / 87.26%
@ Miami / +19 / 8.84%
BC / -6.5 / 67.95%
UNC / +19.5 / 8.47%
@ UVA / -9.5 / 76.08%
@ Clemson / +19 / 8.84%
Syracuse / +12.5 / 18.05%
UGA / +35 / 0.53%

Projected Win Total of 3.97 wins.

Furthermore, I ran a Monte Carlo simulation w/ 10000 iterations to get a distribution of wins. The STDEV is very low (most are in the 1.2, 1.3 range) due to lack of "toss up" games on the schedule, so I don't expect a lot of variance from these numbers.

Georgia Tech	ACC	AVERAGE:	3.9633​		# WINS	FREQ	%
		MODE:	4​		0​	3​	0.03%​
		STDEV:	1.06455​		1​	57​	0.57%​
		MIN:	0​		2​	665​	6.65%​
		MAX:	8​		3​	2557​	25.57%​
					4​	3821​	38.21%​
					5​	2209​	22.09%​
					6​	578​	5.78%​
					7​	102​	1.02%​
					8​	8​	0.08%​
					9​	0​	0.00%​
					10​	0​	0.00%​
					11​	0​	0.00%​
					12​	0​	0.00%​

 

[Northeast Stinger]

SP+ was updated today. As expected, GT improved- from -5.4 to -4.6

23.7 (75) for Offense, 28.2 (79) Defense, and -.1 (115) for Special Teams

You can then project out Regular Season Win totals by taking the difference in opponents SP+ ratings and factoring in Home Field Advantage to get a projected point spread.

Once I get my projected point spread I then convert to the Money Line by using a calculator powered by a ton of historical data located behind a paywall on the Unabated site. Convert the Money Line to implied probability, then sum up the probability for the projected win totals.

Here is what I have after week 1:
Louisville (neutral site) / Loss / 0%
SC State / -28.5 / 97.53%
@ Ole Miss / +24 / 4.64%
@ Wake / +11.5 / 19.23%
Bowling Green / -16 / 87.26%
@ Miami / +19 / 8.84%
BC / -6.5 / 67.95%
UNC / +19.5 / 8.47%
@ UVA / -9.5 / 76.08%
@ Clemson / +19 / 8.84%
Syracuse / +12.5 / 18.05%
UGA / +35 / 0.53%

Projected Win Total of 3.97 wins.

Furthermore, I ran a Monte Carlo simulation w/ 10000 iterations to get a distribution of wins. The STDEV is very low (most are in the 1.2, 1.3 range) due to lack of "toss up" games on the schedule, so I don't expect a lot of variance from these numbers.

Georgia Tech	ACC	AVERAGE:	3.9633​		# WINS	FREQ	%
		MODE:	4​		0​	3​	0.03%​
		STDEV:	1.06455​		1​	57​	0.57%​
		MIN:	0​		2​	665​	6.65%​
		MAX:	8​		3​	2557​	25.57%​
					4​	3821​	38.21%​
					5​	2209​	22.09%​
					6​	578​	5.78%​
					7​	102​	1.02%​
					8​	8​	0.08%​
					9​	0​	0.00%​
					10​	0​	0.00%​
					11​	0​	0.00%​
					12​	0​	0.00%​

I have to admit that this is impressive work.

But to quote the great philosopher mathematician Blaise Pascal, “The heart has reasons which the mind cannot reason.”

 

[Thwg777]

SP+ was updated today. As expected, GT improved- from -5.4 to -4.6

23.7 (75) for Offense, 28.2 (79) Defense, and -.1 (115) for Special Teams

You can then project out Regular Season Win totals by taking the difference in opponents SP+ ratings and factoring in Home Field Advantage to get a projected point spread.

Once I get my projected point spread I then convert to the Money Line by using a calculator powered by a ton of historical data located behind a paywall on the Unabated site. Convert the Money Line to implied probability, then sum up the probability for the projected win totals.

Here is what I have after week 1:
Louisville (neutral site) / Loss / 0%
SC State / -28.5 / 97.53%
@ Ole Miss / +24 / 4.64%
@ Wake / +11.5 / 19.23%
Bowling Green / -16 / 87.26%
@ Miami / +19 / 8.84%
BC / -6.5 / 67.95%
UNC / +19.5 / 8.47%
@ UVA / -9.5 / 76.08%
@ Clemson / +19 / 8.84%
Syracuse / +12.5 / 18.05%
UGA / +35 / 0.53%

Projected Win Total of 3.97 wins.

Furthermore, I ran a Monte Carlo simulation w/ 10000 iterations to get a distribution of wins. The STDEV is very low (most are in the 1.2, 1.3 range) due to lack of "toss up" games on the schedule, so I don't expect a lot of variance from these numbers.

Georgia Tech	ACC	AVERAGE:	3.9633​		# WINS	FREQ	%
		MODE:	4​		0​	3​	0.03%​
		STDEV:	1.06455​		1​	57​	0.57%​
		MIN:	0​		2​	665​	6.65%​
		MAX:	8​		3​	2557​	25.57%​
					4​	3821​	38.21%​
					5​	2209​	22.09%​
					6​	578​	5.78%​
					7​	102​	1.02%​
					8​	8​	0.08%​
					9​	0​	0.00%​
					10​	0​	0.00%​
					11​	0​	0.00%​
					12​	0​	0.00%​

Nice analysis on projections. I have to admit, I predicted 4-8 prior to week 1 and am now leaning to more likely 5-7 after week 1. I think UVA and BC are looking more winnable… BC must have stolen our playbook by losing to NIU week 1.

I’m surprised by the Wake Forest projected line. I have that as WF -4 if played tomorrow. I think that one is the next most winnable conference game after BC and UVA.

My suspicion for the simulations is that it’s based on the odds being static and independent. In reality, things can change a good bit based on team performance over time (hard to gauge after just a week if performance deltas are due to flat out miss reading a team versus statistical week-to-week performance variations).

I don’t have data to pull to see the long-term variations of team preseason win/loss o/u to actual win/loss season records. I’d imagine those have more variation but the betting o/u’s are usually slightly biased to offer worse odds on overs since many square bettors take the over on their team as a means to have ‘action’ all season.

 

[GetYourBuzzOn]

Nice analysis on projections. I have to admit, I predicted 4-8 prior to week 1 and am now leaning to more likely 5-7 after week 1. I think UVA and BC are looking more winnable… BC must have stolen our playbook by losing to NIU week 1.

I’m surprised by the Wake Forest projected line. I have that as WF -4 if played tomorrow. I think that one is the next most winnable conference game after BC and UVA.

My suspicion for the simulations is that it’s based on the odds being static and independent. In reality, things can change a good bit based on team performance over time (hard to gauge after just a week if performance deltas are due to flat out miss reading a team versus statistical week-to-week performance variations).

I don’t have data to pull to see the long-term variations of team preseason win/loss o/u to actual win/loss season records. I’d imagine those have more variation but the betting o/u’s are usually slightly biased to offer worse odds on overs since many square bettors take the over on their team as a means to have ‘action’ all season.

Wake surprised me too. Their SP+ improved from 6.8 to 9.4 after week 1, the rankings must have liked the box score vs. Elon. We'll get a better read on them after they play Vandy, but I still think Wake will be favored by 6+ vs. GT.

PFF pushed out their power rankings today, and they make Wake -13 on a neutral site. I think PFF is underrating GT by 1.5; if that's the case it puts the lookahead w/out injuries weather travel etc at Wake -12, which is in line w/ SP+.

 

[takethepoints]

Wake surprised me too. Their SP+ improved from 6.8 to 9.4 after week 1, the rankings must have liked the box score vs. Elon. We'll get a better read on them after they play Vandy, but I still think Wake will be favored by 6+ vs. GT.

PFF pushed out their power rankings today, and they make Wake -13 on a neutral site. I think PFF is underrating GT by 1.5; if that's the case it puts the lookahead w/out injuries weather travel etc at Wake -12, which is in line w/ SP+.

I never expected us to go into the Wake game as a favorite and I'm surprised that anyone else did. It's away against a team that is averaging 7 - 8 wins a year.

Mind, I think we could beat them with a decent D game. Their OL is as doubtful as ours. Good D however.

 

[billga99]

I am also surprised Syracuse is that big a favorite playing in Atlanta. I would think that would be more of a tossup.

 

[GetYourBuzzOn]

I am also surprised Syracuse is that big a favorite playing in Atlanta. I would think that would be more of a tossup.

Yeah I would be surprised if that was more than SYR -6. The rankings seem to be pretty impressed with their 65-0 win vs Colgate in wk 1.

 

[leatherneckjacket]

SP+ was updated today. As expected, GT improved- from -5.4 to -4.6

23.7 (75) for Offense, 28.2 (79) Defense, and -.1 (115) for Special Teams

You can then project out Regular Season Win totals by taking the difference in opponents SP+ ratings and factoring in Home Field Advantage to get a projected point spread.

Once I get my projected point spread I then convert to the Money Line by using a calculator powered by a ton of historical data located behind a paywall on the Unabated site. Convert the Money Line to implied probability, then sum up the probability for the projected win totals.

Here is what I have after week 1:
Louisville (neutral site) / Loss / 0%
SC State / -28.5 / 97.53%
@ Ole Miss / +24 / 4.64%
@ Wake / +11.5 / 19.23%
Bowling Green / -16 / 87.26%
@ Miami / +19 / 8.84%
BC / -6.5 / 67.95%
UNC / +19.5 / 8.47%
@ UVA / -9.5 / 76.08%
@ Clemson / +19 / 8.84%
Syracuse / +12.5 / 18.05%
UGA / +35 / 0.53%

Projected Win Total of 3.97 wins.

Furthermore, I ran a Monte Carlo simulation w/ 10000 iterations to get a distribution of wins. The STDEV is very low (most are in the 1.2, 1.3 range) due to lack of "toss up" games on the schedule, so I don't expect a lot of variance from these numbers.

Georgia Tech	ACC	AVERAGE:	3.9633​		# WINS	FREQ	%
		MODE:	4​		0​	3​	0.03%​
		STDEV:	1.06455​		1​	57​	0.57%​
		MIN:	0​		2​	665​	6.65%​
		MAX:	8​		3​	2557​	25.57%​
					4​	3821​	38.21%​
					5​	2209​	22.09%​
					6​	578​	5.78%​
					7​	102​	1.02%​
					8​	8​	0.08%​
					9​	0​	0.00%​
					10​	0​	0.00%​
					11​	0​	0.00%​
					12​	0​	0.00%​

I find it hard to believe that Wake and Cuse are that much better than us. Not saying we will beat them, but are they really that good compared to us?

 

[TheJuiceIsLoose404]

At this point in the season SP+ is very heavily skewed to prior years performance. For example, we’re currently ranked above Colorado in SP+ despite them looking like a much better team than us so far in 2023. I would expect that our SP+ will improve over the next few games because I believe we are a significantly better team than the last few seasons.

 

[Heisman's Ghost]

I have to admit that this is impressive work.

But to quote the great philosopher mathematician Blaise Pascal, “The heart has reasons which the mind cannot reason.”

Only on a Tech forum would a Pascal quote be mentioned. The odds of something similar happening on say the Miami or Clemson forum is about the same as the moon falling out of the sky.

 

[Heisman's Ghost]

I find it hard to believe that Wake and Cuse are that much better than us. Not saying we will beat them, but are they really that good compared to us?

They might be but they are not really any better than Louisville. A couple of timely turnovers and a hot quarterback can make a difference in a game against either of them whereas against UGA those things would be a minor annoyance.

 

[GTpdm]

Only on a Tech forum would a Pascal quote be mentioned. The odds of something similar happening on say the Miami or Clemson forum is about the same as the moon falling out of the sky.

Well, to be honest, the Moon is falling out of the sky. Fortunately for us, it has just enough lateral (i.e. orbital) motion to continually miss the Earth as it falls.

 

[Heisman's Ghost]

Well, to be honest, the Moon is falling out of the sky. Fortunately for us, it has just enough lateral (i.e. orbital) motion to continually miss the Earth as it falls.

I will never get used to geek engineer humor. OK, I concede the point. Maybe the Duke revival has shaken the very foundations of the Milky Way after all.