GT should be ranked in the top 25

[AE 87]

I think it just goes to show how close we have been. Sadly though, once you get to the lower ranks, record starts to count for a lot. There shouldn't really have been any 4 loss ranked teams last week, and I think it's a stretch there should be any this week. I think your ppddiff tool is good at estimating how good a team is, but it does not account for how good of a season a team has. If you blow out 6 teams and then lose by 1 the remainder, your ppddiff will be good, but you will still at the end of the day be 6-6.

Wait ... my stat doesn't change the results of the games ... ?

Seriously, though, you're absolutely right that the debate between "Best Team" and "Champion" remains. However, I think your last line in wrong. When you lose a lot of games by one, your PPD Diff and PPD Ratio will be seriously affected.

@bobongo the importance of stats depends on the questions being asked. I appreciate that some fans only care about W-L and that it's a primary factor in ranking polls. However, as others have noted, there are 3 loss and 4 loss teams being ranked, so this is a fairly straight forward way of trying to guage the relative strength of teams with losses.

Again, I appreciate the response that says it wouldn't be an issue if we had taken care of business against SF, Pitt, and Duke like we should have. So, take it for what it's worth.

 

[ibeattetris]

At the end of the day, the only stat that matters is the record, and who that record is compiled against. We don't belong in the top 25, but if we beat Virginia and Georgia (barely within the realm of possibility), we'll be around #15 - #17.

Explain Miss St then? If it was ONLY about your record, they should not be ranked. So there has to be something else as well.

Like I said. No 6-4 team should be ranked, including us, because record is important. But it is not everything.

 

[Whiskey_Clear]

Explain Miss St then? If it was ONLY about your record, they should not be ranked. So there has to be something else as well.

Like I said. No 6-4 team should be ranked, including us, because record is important. But it is not everything.

S
E
C

 

[ibeattetris]

When you lose a lot of games by one, your PPD Diff and PPD Ratio will be seriously affected.

Apart from Clemson, isn't this almost exactly us? In games we are winning, we are dominating ppddiff, but in our losses we have been really close? Of course it will strong influence it, but I think we are proof that losses don't pull you down as much when your wins are solid and your losses are close.

Regardless, I definitely really like this metric, and I am certainly not questioning its validity. If I had to rank a top 25, I would use a combination of ppddiff and pppdiff with some manual shifting due to record.

 

[GTNavyNuke]

I think losing to South Florida and Pitt set us way back in the eyes of folks that vote in these polls. Beating UVA will get us some votes. Then you have to play Georgia and if we lose there we will lose those newly acquired votes. The best thing for us is to win against Virginia and win our bowl game. We'll get some votes and might barely crack the top 25, but realistically we'll probably be somewhere in the top 30 teams. Now if we beat Georgia, forget everything I wrote.

Agree the losses to all FCS teams with winning records put us back. If we beat UVa, the pollsters will look at us being home and the UVa DL being ******. Maybe if we hang 600 yards on them and the D pitches a shutout and ST makes no major mistakes, we'll get credit.

I really don't care about the rankings; we're going to a bowl game which is much much better than I expected after the Duke loss.

 

[bobongo]

Explain Miss St then? If it was ONLY about your record, they should not be ranked. So there has to be something else as well.

Like I said. No 6-4 team should be ranked, including us, because record is important. But it is not everything.

Mississippi State shouldn't be there.

I should have made clearer that I was talking about what the rankings should be about. It should be about your record and your strength of schedule - nothing else. Those are the only two factors that should be considered in the rankings, IMO.

 

[MikeJackets1967]

Mississippi State shouldn't be there.

I should have made clearer that I was talking about what the rankings should be about. It should be about your record and your strength of schedule - nothing else. Those are the only two factors that should be considered in the rankings, IMO.

Mississippi State in my opinion is a better team than 9-1 Utah State who is #14

 

[stech81]

Mississippi State in my opinion is a better team than 9-1 Utah State who is #14

Maybe

 

[ibeattetris]

Mississippi State shouldn't be there.

I should have made clearer that I was talking about what the rankings should be about. It should be about your record and your strength of schedule - nothing else. Those are the only two factors that should be considered in the rankings, IMO.

I think the problem is defining strength of schedule. I like to use quantitative analysis of teams to help determine this. Right now, anyone who beat Miss St looks good because they beat a ranked team. So the ranks then just justify themselves. Using statistics that remove win/loss from the equation can determine how strong a team is, which then can determine the strength of individual wins.

There are hundreds of ranking systems out there, and I like the approaches that combine them as I don't think one is inherently "correct".

 

[cuttysark]

Here's what's more important than being ranked at the moment: Miami was leading the country on defense by holding opponents to 22.1% on converting 3rd downs. GT went 9 for 16 on Saturday night against a defense that NFL Scouts said have 7 to 8 guys who will be drafted next April.

Add to that the fact after Miami scored to make it 27 to 21 with just under 7 minutes left in the game, the Yellow Jackets converted 3rd downs, moved the ball, and the Canes never got it back! Impressive!!

When he was at Georgia Southern, CPJ said: "Nobody wants to see the option in November when it's working." Right now it's working Big Time! That's much more important than what sportswriters think.

 

[Oakland]

Agree the losses to all FCS teams with winning records put us back. If we beat UVa, the pollsters will look at us being home and the UVa DL being ******. Maybe if we hang 600 yards on them and the D pitches a shutout and ST makes no major mistakes, we'll get credit.

I really don't care about the rankings; we're going to a bowl game which is much much better than I expected after the Duke loss.

The point I was trying to make was our team started out slowly after a not so great 2017 season. Being 1-3 at the beginning of the season probably planted a seed in the pollsters minds that this was not going to be one of GT's better years. Hard work and not giving up is starting to show that this team can compete. Like it or not rankings do help. People need to see progress and that helps with donations, media attention, and the big one...ticket sales/attracting new fans.

 

[Whiskey_Clear]

Mississippi State in my opinion is a better team than 9-1 Utah State who is #14

I’m gonna have to dig back through the archives to see if you said similar about UCF v Auburn before their bowl game

 

[bobongo]

I think the problem is defining strength of schedule. I like to use quantitative analysis of teams to help determine this. Right now, anyone who beat Miss St looks good because they beat a ranked team. So the ranks then just justify themselves. Using statistics that remove win/loss from the equation can determine how strong a team is, which then can determine the strength of individual wins.

There are hundreds of ranking systems out there, and I like the approaches that combine them as I don't think one is inherently "correct".

There is no perfect solution, because every team only plays a tenth of the other teams out there. I see your point, but removing wins/losses from the equation? You lost me there. W's and L's are what it's all about. That's why you play the games, to find out who wins, not who runs up the most yardage. It's an endless argument, but it won't be solved by putting a template of useless stats over everything.

 

[tech_wreck47]

After going on this winning streak the losses hurt even more now. I’m just sitting here thinking what this season would be like if we had beaten USF, Pitt, and even Duke like we were very capable of doing so. 9-1 with a loss to Clemson and we would be a top 10 team probably. Really shows you ho football is a game of inches.

 

[dressedcheeseside]

Are any of our wins against an opponent that will have a record above .500? VT and MIA might get there, but we haven’t shown any reason to be a top 25 team so far. I’d agree 20-25 would be the CEILING if we win out...

Our loss to Clemson was better than anybody else’s.

 

[dressedcheeseside]

A top 25 caliber team would have no more than two losses with this schedule.

We only have maybe three decent teams on our schedule. Preseason people were touting this crappy schedule as a real murderers row...

And yet every team on our schedule save Alcorn outrecruits us every year.

 

[tech_wreck47]

And yet every team on our schedule save Alcorn outrecruits us every year.

Not only that, but our schedule isn’t as easy as people are making it out to be and i don’t think we would even have a top 25 if all teams with a schedule like ours could only have 2 losses.

 

[bobongo]

Not only that, but our schedule isn’t as easy as people are making it out to be and i don’t think we would even have a top 25 if all teams with a schedule like ours could only have 2 losses.

Oh, now. If we had only two losses, we'd easily be in the top 20.

 

[GTJake]

IMO, ranking outside the Top10 doesn't mean a whole lot right now, but getting a Top25 ranking in the final poll gets us some off-season PR, a little recruiting bump and some momentum going into 2019 ...

 

[ibeattetris]

There is no perfect solution, because every team only plays a tenth of the other teams out there. I see your point, but removing wins/losses from the equation? You lost me there. W's and L's are what it's all about. That's why you play the games, to find out who wins, not who runs up the most yardage. It's an endless argument, but it won't be solved by putting a template of useless stats over everything.

Only when determining "strength" of schedule. Rankings should be some combination of record + strength of schedule. If you include record in each side of that, you are just muddying it. For example, while we lost to USF, we were easily the better team. Stats will show that, but a binary "win or loss" does not. There is a reason that win/loss elo systems of ranking have never been very good predictors of wins in ncaa, where as point per play differential/point per drive differential does.

I am certainly not saying record doesn't matter, I am saying that performance on the field needs to be considered as well. From what I have heard, the CFP committee agrees somewhat.