Defensive Score Sheet: Kentucky vs Georgia
Here is the defensive score sheet for Saturday's game against Georgia. For an explanation of what this is, you can go here. I revamped the sheets I use to keep track of what happens so in the future I should be better able to answer some of the questions people had last time, especially concerning the offense/defense matchups. I won't have it for this game, but I might be able to do it for Ole Miss and should be able to do so for Florida.
The Georgia came presented a couple of challenges in determining how to classify shots due to the zone that was played in the first half. I think this game in particular had a lot more shots where it wasn't obvious who the defender(s) should be (if any) so I'll be interested to know how this jives with what people saw and remember from the game.
| Kentucky | Shot Defense | ||||||
| Name | Min | Def Rb | Blocks | Forced Miss | Def FGM | FT Miss | Def FTM |
| Doron Lamb | 31 | 2 | 5 | 0.5 | |||
| Terrence Jones | 26 | 7 | 1 | 8.5 | 3 | 2 | 2 |
| TEAM | 0 | 6 | 2 | ||||
| DeAndre Liggins | 36 | 2 | 1 | 4 | 2.5 | 4 | |
| Eloy Vargas | 5 | 0 | 1 | 1 | 0.5 | ||
| Josh Harrellson | 30 | 7 | 2 | 6.5 | 4 | 3 | 4 |
| Brandon Knight | 40 | 4 | 3 | 5 | 1 | 1 | |
| Darius Miller | 32 | 5 | 2 | 2.5 | 1 | 3 | |
| Tracked Totals | 36 | 20 | 7 | 14 | |||
| BoxScore | 27 | 5 | 36 | 21 | 7 | 14 | |
- I've included the box score totals so you can see how they matchup. Somehow I missed an UGA field goal, but otherwise the accuracy in the totals is pretty good.
- Terrence Jones once again was a top performer in defending missed shots. Harrellson was second and Georgia missed a number of shots that were open either by poor defense by Kentucky or good passing from Georgia (or both).
- Lamb had a pretty solid game. He didn't send anyone to the line and did a good job forcing misses.
| Kentucky | Ball Handling | |||
| Name | Min | Steals | F.TO | Deflections |
| Doron Lamb | 31 | 1 | ||
| Terrence Jones | 26 | 1 | 0.5 | 1 |
| TEAM | 3 | |||
| DeAndre Liggins | 36 | 2 | 4 | 7 |
| Eloy Vargas | 5 | |||
| Josh Harrellson | 30 | 3 | 1.5 | |
| Brandon Knight | 40 | 3 | 1 | |
| Darius Miller | 32 | 1 | ||
| Tracked Totals | 13 | 10 | ||
| BoxScore | 6 | 13 | ||
- This just in. DeAndre Liggins is really active defensively. Liggins leads the team in deflections again and forced 4 turnovers. Actually, he had a hand in 5 TO's, forcing 3 by himself and hooking up with a teammate a couple of times.
- You'll notice that I have Josh with only 1.5 forced TOs despite his 3 steals. One of his steals was split credit with another player. The other was a case where the Georgia player basically just threw the ball right to him. In that instance the only thing Josh did was happen to be standing in the right place at the right time, he didn't really do anything to "force" it, so I put that one in the "Team" category.
- Liggins and Knight both had tie-ups that resulted in a jump ball where Georgia kept the ball. I counted those as Deflections since they fall under the general idea of disrupting the offense.
| Kentucky | Calculations | ||||
| Name | Min | Stops | Stop% | %Tm Def. Poss | Def Rtg |
| Doron Lamb | 31 | 3.5 | 0.9 | 0.06 | 85.84 |
| Terrence Jones | 26 | 7.3 | 0.7 | 0.17 | 87.43 |
| TEAM | 4.5 | 0.7 | 0.1 | 87.43 | |
| DeAndre Liggins | 36 | 6.3 | 0.6 | 0.16 | 89.55 |
| Eloy Vargas | 5 | 0.3 | 0.3 | 0.01 | 90.23 |
| Josh Harrellson | 30 | 7.9 | 0.6 | 0.2 | 90.33 |
| Brandon Knight | 40 | 6.4 | 0.5 | 0.18 | 91.64 |
| Darius Miller | 32 | 3.8 | 0.5 | 0.11 | 92.04 |
| Tracked Totals | 39.9 | 0.6 | 0.99 | 87.3 | |
| BoxScore | 39.9 | 0.6 | 1 | 89.24 | |
- Here are the calculated metrics based on the previous two tables. I didn't go into any detail on Defensive Rating last time, but you can think of it as sort of a basketball equivalent to ERA in baseball. It's an estimate of the number of points that are scored against a player, scaled out to 100 possessions (in the same way ERA is scaled out to 9 innings).
- Liggins' rating is basically "team average" which I thought was odd, but it's mostly due to the guys he sent to the line hitting all their free throws.
- You can see that Josh was involved in fully 20% of the defensive plays and personally stopped them 60% of the time. That tracks with the inside oriented approach Georgia used in the game.
- When Lamb was directly involved in the defensive play, he ended it 90% of the time without the opponent scoring. Unfortunately, he was only involved in about 6% of the plays.
Here are the full tables. For Georgia's defense, they have 1 more TO forced than the box score and I think that came on a play where a UK player got a defensive rebound only to have it stripped by a Bulldog. I counted it as a TO while the official scorer counted it as an offensive rebound by Georgia. That's my best guess anyway.
| Kentucky | Shot Defense | Ball Handling | Calculations | |||||||||||
| Name | Min | Def Rb | Blks | F.Miss | Def FGM | FT Miss | Def FTM | Stls | F.TO | Defl | Stops | Stop% | %TmDPoss | Def Rtg |
| Doron Lamb | 31 | 2 | 5 | 0.5 | 1 | 3.5 | 0.9 | 0.06 | 85.84 | |||||
| Terrence Jones | 26 | 7 | 1 | 8.5 | 3 | 2 | 2 | 1 | 0.5 | 1 | 7.3 | 0.7 | 0.17 | 87.43 |
| TEAM | 0 | 6 | 2 | 3 | 4.5 | 0.7 | 0.1 | 87.43 | ||||||
| DeAndre Liggins | 36 | 2 | 1 | 4 | 2.5 | 4 | 2 | 4 | 7 | 6.3 | 0.6 | 0.16 | 89.55 | |
| Eloy Vargas | 5 | 0 | 1 | 1 | 0.5 | 0.3 | 0.3 | 0.01 | 90.23 | |||||
| Josh Harrellson | 30 | 7 | 2 | 6.5 | 4 | 3 | 4 | 3 | 1.5 | 7.9 | 0.6 | 0.2 | 90.33 | |
| Brandon Knight | 40 | 4 | 3 | 5 | 1 | 1 | 3 | 1 | 6.4 | 0.5 | 0.18 | 91.64 | ||
| Darius Miller | 32 | 5 | 2 | 2.5 | 1 | 3 | 1 | 3.8 | 0.5 | 0.11 | 92.04 | |||
| Tracked Totals | 36 | 20 | 7 | 14 | 13 | 10 | 39.9 | 0.6 | 0.99 | 87.3 | ||||
| BoxScore | 27 | 5 | 36 | 21 | 7 | 14 | 6 | 13 | 39.9 | 0.6 | 1 | 89.24 | ||
| Georgia | Shot Defense | Ball Handling | Calculations | |||||||||||
| Name | Min | Def Rb | Blks | F.Miss | Def FGM | FT Miss | Def FTM | Stls | F.TO | Defl | Stops | Stop% | %TmDPoss | Def Rtg |
| Gerald Robinson | 37 | 3 | 6.5 | 2 | 1 | 2 | 2 | 1 | 6.2 | 0.8 | 0.12 | 93.2 | ||
| Marcus Thornton | 11 | 3 | 1 | 2 | 0.5 | 1 | 1 | 3.44 | 0.9 | 0.06 | 94.39 | |||
| Jeremy Price | 24 | 6 | 5 | 2 | 1 | 3.91 | 0.8 | 0.07 | 94.46 | |||||
| Trey Tompkins | 29 | 9 | 4 | 3.5 | 6.59 | 0.7 | 0.15 | 95.47 | ||||||
| Sherrard Brantley | 9 | 0 | 0 | 0.0 | 0 | 98.5 | ||||||||
| Donte Williams | 8 | 2 | 1.5 | 0.5 | 2 | 1.69 | 0.5 | 0.05 | 98.68 | |||||
| Chris Barnes | 20 | 3 | 1 | 2 | 1.71 | 0.5 | 0.05 | 99.61 | ||||||
| Travis Leslie | 31 | 0 | 2 | 3 | 1 | 2 | 2 | 2 | 2.84 | 0.4 | 0.1 | 101.75 | ||
| Dustin Ware | 31 | 2 | 7 | 5.5 | 1 | 1 | 1 | 3.81 | 0.4 | 0.15 | 104.22 | |||
| TEAM | 1 | 6 | 6 | 1 | 7 | 5 | 7.87 | 0.5 | 0.25 | 104.34 | ||||
| Tracked Totals | 29 | 22 | 9 | 16 | 10 | 3 | 38.06 | 0.6 | 1 | 98.12 | ||||
| BoxScore | 29 | 1 | 29 | 22 | 9 | 16 | 5 | 9 | 37.06 | 0.6 | 0.99 | 98.5 | ||
Update (2-07-2011) I discovered an error in the calculations that affected the calculated values. The correct tables are below.
| Kentucky | Shot Defense | Ball Handling | Calculations | |||||||||||||
| Name | Min | Def Rb | Blks | F.Miss | Def FGM | FT Miss | Def FTM | Stls | F.TO | Defl | Stops | DScPoss | DPoss | Stop% | %TmDPoss | Def Rtg |
| Terrence Jones | 26 | 7 | 1 | 8.5 | 3 | 2 | 2 | 1 | 0.5 | 1 | 7.1 | 3.9 | 10.97 | 0.6 | 0.25 | 86.72 |
| Josh Harrellson | 30 | 7 | 2 | 6.5 | 4 | 3 | 4 | 3 | 1.5 | 7.4 | 5.8 | 13.18 | 0.6 | 0.26 | 91.46 | |
| Darius Miller | 32 | 5 | 2 | 2.5 | 1 | 3 | 1 | 3.1 | 3.9 | 6.9 | 0.4 | 0.13 | 93.85 | |||
| Brandon Knight | 40 | 4 | 3 | 5 | 1 | 1 | 3 | 1 | 6.0 | 5.5 | 11.46 | 0.5 | 0.17 | 92.15 | ||
| Doron Lamb | 31 | 2 | 5 | 0.5 | 1 | 3.8 | 0.5 | 4.33 | 0.9 | 0.08 | 84.25 | |||||
| DeAndre Liggins | 36 | 2 | 1 | 4 | 2.5 | 4 | 2 | 4 | 7 | 6.4 | 4.3 | 10.74 | 0.6 | 0.18 | 89.32 | |
| Eloy Vargas | 5 | 0 | 1 | 1 | 0.5 | 0.4 | 0.5 | 0.89 | 0.4 | 0.11 | 93.18 | |||||
| TEAM | 40 | 0 | 6 | 2 | 3 | 5.4 | 2.0 | 7.35 | 0.7 | 0.11 | 86.33 | |||||
| Tracked Totals | 200 | 27 | 5 | 36 | 20 | 7 | 14 | 6 | 13 | 10 | 39.5 | 26.3 | 65.8 | 0.6 | 0.2 | 89.25 |
| BoxScore | 200 | 27 | 5 | 36 | 21 | 7 | 14 | 6 | 13 | 39.5 | 27.3 | 66.82 | 0.6 | 0.2 | 89.6 | |
| Georgia | Shot Defense | Ball Handling | Calculations | |||||||||||||
| Name | Min | Def Rb | Blks | F.Miss | Def FGM | FT Miss | Def FTM | Stls | F.TO | Defl | Stops | DScPoss | DPoss | Stop% | %TmDPoss | Def Rtg |
| Trey Tompkins | 29 | 9 | 4 | 3.5 | 4.45 | 3.5 | 7.95 | 0.6 | 0.12 | 98.62 | ||||||
| Jeremy Price | 24 | 6 | 5 | 2 | 1 | 1.61 | 0.9 | 2.51 | 0.6 | 0.04 | 97.84 | |||||
| Travis Leslie | 31 | 0 | 2 | 3 | 1 | 2 | 2 | 2 | 3.49 | 3.9 | 7.39 | 0.5 | 0.11 | 100.79 | ||
| Dustin Ware | 31 | 2 | 7 | 5.5 | 1 | 1 | 1 | 5.34 | 6.0 | 11.29 | 0.5 | 0.17 | 101.98 | |||
| Gerald Robinson | 37 | 3 | 6.5 | 2 | 1 | 2 | 2 | 1 | 7.17 | 2.0 | 9.17 | 0.8 | 0.14 | 91.76 | ||
| Marcus Thornton | 11 | 3 | 1 | 2 | 0.5 | 1 | 1 | 2.95 | 0.5 | 3.45 | 0.9 | 0.05 | 95.12 | |||
| Chris Barnes | 20 | 3 | 1 | 2 | 0.56 | 1.9 | 2.46 | 0.2 | 0.04 | 101.3 | ||||||
| Donte Williams | 8 | 2 | 1.5 | 0.5 | 2 | 1.41 | 1.4 | 2.81 | 0.5 | 0.04 | 99.09 | |||||
| Sherrard Brantley | 9 | 0 | 0 | 0.0 | 0 | 0.0 | 0 | 98.5 | ||||||||
| TEAM | 40 | 1 | 6 | 6 | 1 | 7 | 5 | 9.45 | 9.2 | 18.6 | 0.5 | 0.28 | 102.02 | |||
| Tracked Totals | 29 | 1 | 29 | 22 | 9 | 16 | 5 | 10 | 3 | 36.43 | 29.2 | 65.6 | 0.6 | 0.98 | 100.53 | |
| BoxScore | 200 | 29 | 1 | 29 | 22 | 9 | 16 | 5 | 9 | 35.43 | 29.2 | 64.63 | 0.5 | 0.96 | 98.5 | |
24 comments
|
0 recs |
Do you like this story?
Comments
Gives Me A Headache
And 50 years ago I was a math major. But I read it, muchas gracias.
"I bring reason to your ears, and, in language as plain as ABC, hold up truth to your eyes." Thomas Paine, December 23, 1776
by Wild Weasel on Jan 30, 2011 6:16 PM EST via mobile reply actions
Well, hopefully math isn't needed to understand most of it
I can see it with the calculations portion, but I’m trying to make that as intuitive as possible. The goal is to aid what we already look for in defense – forcing missed shots, forcing turnovers, being active – and trying to present a more complete picture of what happens in a game than what we get from just our own memory.
So for instance, if you watched the UGA game and thought the Bulldogs were scoring a lot against Knight, that would certainly seem to track with what I saw too. OTOH, if it seemed like Lamb wasn’t playing much defense, then I think there is reason to disagree as he did force some missed shots and didn’t send anyone to the line.
3 > 2, except for very large values of 2.
The tie balls
are a half a turnover in a sense when the possessing team keeps the ball because the next time you get the ball.
I woke up feeling BLUE this morning. It's gonna be a great day.
Well, maybe
I can see that, but in terms of actually recording what happens, I think it fits better under deflection. There’s no guarantee that the next jump ball will result in a TO, or even that there will be a next jump ball at all.
3 > 2, except for very large values of 2.
Thanks JL
I must admit I don’t understand all of it yet, but it will grow on me. I appreciate all the time you invest to capture, analyze, and present the data.
Honestly, I don't understand it all either
Recording defense this way is a fairly new idea and it hasn’t been done very much, so I think it is going to be a while before we really know how to read these in the same way we know how to read a typical box score.
One of the reasons for posting these on ASoB is that I hope the thoughtful, intelligent, insightful members will collectively be able to figure out some things about what gets recorded.
3 > 2, except for very large values of 2.
Those adjectives
Leave me out! ’-) Unless FUN is thoughtful?!? j/k
But I do realize and appreciate the effort you put into this JLev. I see it as pioneer-like. Love our intelligent members.
Blue... there is no other color to Bleed !!!
Well, those adjectives were meant to be inclusive, not exclusive
I think everybody here has something they can contribute! But even if it is something like “I thought Liggins was the best defender, but here he is middle of the pack” I think there is value in discussing the observations during a game that lead to that conclusion and comparing it to what I write down.
3 > 2, except for very large values of 2.
I know, I got the j/k
But I’m going to keep pushing for people to offer up their opinions because that’s what makes it interesting.
I’m glad that people appreciate the effort and I know that trying to understand everything can be a bit daunting when the only time you see it is in one of these posts, but what I really want to try and spark is some debate as to whether these counts are matching up with what everyone else is seeing during the game.
3 > 2, except for very large values of 2.
The only problem
with trying to match your system with real game observations has already been pointed out. When we are watching the game it is subjective observation vs objective such as watching replays once the game is over. When its close or something negative is happening to UK, the weight of remembered observation goes to those moments with the most impact like a steal, turn over or a blocked shot. Those memories are the ones that leave us with our game impression rather than the nuts and bolts of good defense.
I woke up feeling BLUE this morning. It's gonna be a great day.
That's definitely a problem, but one that I think can be overcome
3 > 2, except for very large values of 2.
It would take
more concentration than I am able to muster these days.
I woke up feeling BLUE this morning. It's gonna be a great day.
I'm thinking more of a wisdom of crowds approach
If 4 people think Jones had a bad defensive game and 3 people think he had a good game, then between them they would ideally be able to come up with 7-8 instances or specific impressions that could then by compared to what I come up with.
That would give us a better idea of exactly what these sheets do and do not tell us about an individual defensive effort. Clearly we aren’t going to know everything about defense with something like this, but I think we can know a bit more.
3 > 2, except for very large values of 2.
You have my vote
I think it’s fascinating, never bothered with stats before other than boxscore and play by play.
I always looked at nonball movement, rebound positioning, ect. Now I am hooked and can’t get enough. I appreciate every thing you do.
"SPORTS"--Not interested----"CATS"--Pull up a chair,I've got all night.
Great work yet again
Excellent job compiling all of this, JL.
One question regarding these tables: is there a way to somehow indicate what’s a good/bad rating? I.e. for a given column, higher numbers are better, another column lower is better? That way it will be easier for those of us who are, shall we say, less mathematically inclined (such as myself), to better grasp what we are reading? It wouldn’t really require more work every time, just tacking on a “legend” if you will, indicating what we should be looking for.
Just a thought. I admire the effort it takes to record and translate all this. Want to make sure that more people can benefit from what you’ve painstakingly done.
by wildcatfaninexile on Jan 31, 2011 9:03 AM EST reply actions
Clarification
That extra line I was talking about really only applies to some of the “calculations” columns. I think we can all pretty easily tell that getting 7 steals or a dozen deflections is a good thing…. I hope…haha!
by wildcatfaninexile on Jan 31, 2011 9:07 AM EST up reply actions
Sure
Stops: this is the number of time an individual does something that ends the opponent’s offense without any points scored of any kind. This is what you want your defense to do, so higher is better.
Stop%: this is a ratio of the number of stops a player gets compared to the total number of times they are involved in a play that ends the offensive possession. That includes both when the opponent scores and when they do not score, so higher is better.
%TmDPoss: Each game consists of some number of possessions by each team, so there are X number of times when that team is on defense. This is the percentage of those X times that a player is involved with ending the possession. This one doesn’t really have a “good” or “bad” associated with it, but in general we want our best defensive players to be involved in the most number of possessions, so you can think of it that way.
Def Rtg: As I mentioned above, you can think of this in the same way as ERA in baseball – lower is better because this is essentially the number of points a defensive player allows the other team to score.
3 > 2, except for very large values of 2.
Great stuff!
Wonderful, wonderful breakdown. Thanks for taking the time and sharing!
Thanks! Glad you like it!
3 > 2, except for very large values of 2.
This thing is growing on me, JL...
I’m starting to wish I had all the calculations for it – especially the weighting for each stat in the overall rating to better understand why my impressions of the game don’t quite align with individual players’ ratings. I’m sure much of that is viewing bias/emotional investment while watching the game, but I imagine the originator’s weightings on various defensive skills and my own are also varied.
I'll post the formulas in the next sheet
I am with you – I like to know how calculations are arrived at and what goes into them. I’ll post the formulas along with a brief explanation for some of the parts that Oliver wrote about in his book.
Glad to hear it’s growing on you!
3 > 2, except for very large values of 2.

by 








