ADVERTISEMENT

Defensive Metrics

KA4Prez

All-American
Dec 8, 2003
17,982
4,297
113
41
Columbus, OH
Don't get caught up on them too much. The reality is, 12 of the top 40 offenses are in the SEC. The defense is going to be difficult to improve in almost every game we play. It doesn't mean UK can't improve or doesn't have flaws, but all of the "defensive baselines" are likely going to say that only Auburn, and maybe Tennessee, if their offense improves, can win the title from the SEC, based on historical trends. However, that won't coincide with the fact that the SEC likely has 6 or 7 teams who are seeded on the 5-line or higher. The math won't end up mathing from a probability standpoint.
 
A few thoughts

1. We've actually improved quite a bit. IIRC, at 1 point, we were 97th in defensive efficiency on Kenpom. We're at 75 today.

2. The whole "Champions finish in the Kenpom top 30 on offense and defense" only applies after they've won the championship. It's amazing what beating 6 consecutive tournament level teams will do for your metrics.

3. Bad defensive teams have made the Final 4. Alabama finished 111th in Kenpom defenisve efficiency last year in spite of winning 4 games. Miami finished 99th in defensive efficiency on kenpom in 2023 when they had their run to the Final 4.

So you CAN overcome the defense. You need to improve the defense as much as you can obviously. Rebounding becomes all the more crucial. And then you just have to be able to execute your offense.

I'm not booking a hotel or anything. But I agree--we shouldn't feel doomed over the defensive efficiency numbers.
 
The metrics are adjusted for this already. Us playing good offenses is fully taken into account when the defensive efficiency metric is calculated. You shouldn't follow the metrics 100% and they aren't perfect, but they are all adjusted for your opponent strength.

I understand.

What I'm saying is that it's going to be very difficult for Kentucky to make up ground this late in the season, given the SEC's strength.

And I'm also saying that the numbers will say that only 1 or 2 SEC teams have a shot to win the title.
Yet they will likely have 7 or 8 of the top 24 seeded teams.

Those numbers will conflict, as every Final Four/National Title team likely comes from that range.
 
Makes no sense, the metrics compensate for the teams you are playing.

I would argue that we have an advantage in that we are playing against better offenses, so if we can hold them below their averages, we will rise in the rankings.

We get Alabama this Saturday, if we can keep them in the 70's, our metrics will go up.
 
They are nerd numbers. No one cares. KP nerd numbers are even fake nerd numbers. He removes possessions to figure offensive efficiency.

I wish KP was a coach. For our opponents.
 
And how are they ranked so high
After you posed the question I looked and our next 8 opponents (just picked the next 8 because that was all that I could jot down on a sticky pad) have a combined record of 104-27 with an average NET ranking of 32 and Duke's next 8 opponents have a combined record of 82-51 with an average NET ranking of 106.
 
  • Like
Reactions: UK67KU59
Not all that worried. It's a game that favors offense and Pope knows this. There is a path where if your offense is so good that it out paces them, than the defensive issues might not matter.

I think as long as we can stay in that 70-50 range, we should be fine. If we drop from here to near the 100s or beyond, that might be a problem (as it also means some teams will have had to have a field day against us, and likely won).

We just played 3 top15 games in 4 days, this is probably our hardest stretch of the SEC conference play, with the next 3 games being right there (TAM, Bama, Vandy, Tenn, but we already got 1 win there). I expect our defense to make some stronger showings when we get Texas, Vandy, LSU, Oklahoma, Mizz. Those games won't be easy.. but they won't be as bad as this early stretch.

Take this for example.. When we play Texas on February 15th, we will only have 2 ranked teams left out of 7 games we have left. This is seemingly the easiest 1/3rd of the schedule.
 
Last edited:
Missouri could be ranked next week if they beat the Razorbacks Saturday.
After you posed the question I looked and our next 8 opponents (just picked the next 8 because that was all that I could jot down on a sticky pad) have a combined record of 104-27 with an average NET ranking of 32 and Duke's next 8 opponents have a combined record of 82-51 with an average NET ranking of 106.
Did a little more research and up to this point Duke's opponents have an average record of 9.76-6.64 for a 57% win percentage. Their opponents average net ranking is 102.68. On the other hand UK's average opponent record is 10.23-6.17 for a 60% win percentage and UK's opponents average net ranking is 110.82. We are close now but those number will drastically swing in UK's favor playing in the SEC against quality opponents as opposed to Duke playing in ACCrappy conference.
 
Last edited:
Makes no sense, the metrics compensate for the teams you are playing.

I would argue that we have an advantage in that we are playing against better offenses, so if we can hold them below their averages, we will rise in the rankings.

We get Alabama this Saturday, if we can keep them in the 70's, our metrics will go up.

They will go up, but it's nearly impossible for Kentucky to crack, like the top 50 in defensive efficiency. Which will lead to people saying they "can't win the national title."
 
The metrics are adjusted for this already. Us playing good offenses is fully taken into account when the defensive efficiency metric is calculated. You shouldn't follow the metrics 100% and they aren't perfect, but they are all adjusted for your opponent strength.
This is absolutely true, but how accurate that method actually is can be a point of contention. For as long as I can remember having computer metrics you have had people saying they rate teams who don't play anyone too highly. Just because there is a system in place to do a certain thing doesn't necessarily mean it does that thing optimally or even well.
 
ADVERTISEMENT
ADVERTISEMENT