You need both. Our good NET probably allows us to make it with only 3 Q1 wins. Last year, we needed all 4. There's a chance we could get in with only 2 Q1s but probably not.
Netscape - Daily discussion on scores and Net impact
- Thread starter Dointhatrag
- Start date
You need both. Our good NET probably allows us to make it with only 3 Q1 wins. Last year, we needed all 4. There's a chance we could get in with only 2 Q1s but probably not.
We got blown out a few times in the OOC, and had a bunch of other OOC games where we won against really bad teams, but looked bad in the process (William and Mary, Alabama State). We did not perform how the model expects a tournament team to perform against bad opponents.
This year, we aren’t getting blown out at all. And we’re doing what the model expects for a good team to do against bad teams. We need to get some resume chips, but the model likes what it’s seen from us…which it never did last year.
At Cuse, at wake, and at Bc are all high Q2 that could be Q1. That’s not murderers row. This team should have 4-5 when all is said and done if they plan to make the tourney.
Maybe one of those will end up Top 75. Probably not all 3. And its more likely than not that none of them will.
Blowouts aren't the cause, they are the effect.
The NET rewards teams that are efficient. Or in other words, teams that play well. And teams that play well tend to blow out bad teams.
Blowouts and good efficiency are the exact same thing. Stop defending NET. You remind me of an NRA lobbyist. Everyone hates NET.
I'll make you a deal. The day that you actually figure out what the NET is and how it is used I'll stop trying to explain it to you.
I don't anticipate that day coming any year soon.
Are you saying a great efficiency game and a blowout aren't exactly the same thing?
Michigan St jumped Pitt last night moving from 42 to 33 with a 99-55 win over stony brook. Yes blowouts matter.
BTW Mich st in now 7-5 with that win going 1-5 in Q1 games
BTW Mich st in now 7-5 with that win going 1-5 in Q1 games
I'm saying exact what I said a couple posts ago. You don't understand cause and effect. Playing well, playing efficient, causes blowouts. Blowouts don't cause playing efficient. Blowouts are the effect, not the cause.
It's like you have no idea what they are trying to measure. You think they are measuring blowouts. They are measuring efficiency. Efficiency is nothing more than a word that means playing well. Teams that play well are, by definition, efficient. Teams that don't play well, by definition, are not efficient.
That's all they are doing, measuring who is playing well and who is not, and then ranking the teams by who has played well and who has not. Nothing more, nothing less.
But you understand how this discourages teams from playing walk-ons to win by wider margins for instance, right?
NET is similar to RPI which is a good tool for gamblers. Efficiency is a good predictive indicator. But I don't think a team that takes their foot off the gas for sportsmanship should be penalized. That tweak would be good. And don't tell me they wouldn't tweak it for a social goal because they already increased the bonus for road wins to encourage top teams to actually play some road games.
NET is not similar to RPI. And if you really think that it is you either have no idea what NET is, what RPI was, or both.
I would have no problem if they put some sort of limiting factor on the rating in big wins. But you'd have to do a lot of thinking, certainly a lot more than SMF has put into it with his ridiculous idea on the topic, to figure out the best way to do it.
For example, on Wednesday night Pitt was ahead by 24 points with just over five minutes to go in the game. So do you call that a blowout at that point? But then what about the fact that Pitt did not, in fact, put in the walkons or backups (Hueitt was actually at the scorers table to come in at one point, but got called back). And yet over those last five minutes the score went from 24 to 14. Are you going to count that or not?
And then the opposite situation, against SC State we were ahead by 28 with just under five minutes left when we cleared the bench. And the bench guys stretched the lead out to as many as 38 before we ended up winning by 36. So playing the bench guys not only did not hurt Pitt's efficiency, they actually almost certainly helped. Are you going to count that or not?
You dont count anything once the game is determined to be over by some mathematical calculation. Like if Kansas takes a 2-0 lead over South Carolina State. Done. Game doesn't count for efficiency purposes. Doesnt matter if Kansas wins by 1 or 100. This is an exaggeration but not a huge one. It would have to be something similar. It would also encourage a Kansas to use bench guys more and rest starters knowing they dont have to win by 50
I meant Pomeroy not RPI. I listened to the interview with Ken about NET and his system and he explained why they're roughly similar. But both measure predictively more than a resume comparison based mostly on SoS + wins
I told you before why that would be silly, but since you don't understand math you don't get it.
Do you really think it is that impossible for some mathematical formula to determine when a game is not winnable and then not count possessions past that point?
No, I think that if you cut off counting a game after one team scores a basket on the first possession of the game you have made that team's offensive and defensive efficiencies for the game through the roof, way, way better than anyone ever would be over the course of a whole game.
And I explained that to you before, but as I said, since you don't understand math you don't get it. At all.
Yes, you have made it go through the roof for that 1 game but efficiency isnt weighted on a per game average. Its weighted on a per possession average. So the 1 Kansas possession where they scored on SC St will be just 1 of 1000 or so possessions on the season. The fact that they were 1/1 for 100% in that game will not matter as much as winning by 60 or winning by 1.
If you can think that why do you simultaneously think that four or five possessions by walkons at the end of a game kill a team's NET for the season?
You get that those are opposite points of view, right?
Oh, that's right, you don't.
It doesn't kill it for a season. But if a 30 point lead becomes 20 on the the last 4 or 5 possessions, it hurts you to a degree. That 1 time probably only costs you 2 spots in NET. If your walk-ons keep getting outscored by 10 at the end of the game then your NET will be much lower than if they hadnt.
Stick with me and you'll get it.
But of course here in the real world that isn't what actually happens. I mean just two games ago we put our walkons in the game and got MORE efficient, not less. Sometimes you put your walkons in and they make no real difference at all. And other times, like our last game, you don't put your walkons in and the other team actually plays better than you at the end.
If the only thing that ever happened was that every game you put your walkons in for the last two minutes of the game and your walkons always sucked then yeah, that would affect your ranking by a little bit. But that isn't what actually happens in the real world. As anyone who pays any attention knows.
Obviously.
But the point is that you need to keep your foot on the gas even after the game has been decided because NET values each possession the same. It values a possession in a 72-70 game the same as in a 104-45 game. So, to maximize NET, you want your best players to play as long as possible even in blowouts even though yes, its possible you can take them out and put in walk-ons and extend your lead.
And yet you keep bringing up the "you put your walkons in at the end of the game and you get outscored by ten points and your NET drops because of it" argument all the time. Which certainly at least implies that it's not obvious to you.
Do you really think I mean that its an automatic? Its an example. NET shouldn’t be calculating efficiency when the game has been decided. Everyone here but you agrees with me on that.
Joe and smf it is clearly obvious neither of you know nothing of the NET please take your drivel to pm.
This ought to be funny.
Just what is it that you don't think I understand about it?
Helps for the NIT. Need to stay in the ACC Top 4 of NET for teams who dont make it.
As of 1/4 Needless to say the game against Louisville is a must win and even then will drop us a spot or two.
12 North Carolina
16 Clemson
20 Duke
40 Miami (FL)
49 Virginia
52 Pittsburgh
55 Wake Forest
65 Virginia Tech
71 NC State
78 Syracuse
81 Boston College
123 Georgia Tech
131 Florida St.
181 Notre Dame
266 Louisville
12 North Carolina
16 Clemson
20 Duke
40 Miami (FL)
49 Virginia
52 Pittsburgh
55 Wake Forest
65 Virginia Tech
71 NC State
78 Syracuse
81 Boston College
123 Georgia Tech
131 Florida St.
181 Notre Dame
266 Louisville
Miami played their way into the field for the moment by beating Clemson last night. I still think the ACC will get 5. Probably UVa because their defense will win them enough games. But that 5th spot will come from 1 of UVa, Pitt, Wake, VT, NC St, Syr, BC. Keep an eye on Wake as they are surging. If we can go 3-4 now vs Duke and then road games at Duke, UVa, Miami, Clem, Wake, NC St, we can get in if we go 10-0 in the Q2-4, with 8 of them being at home
The RPI worked that way, but the NET does not. If Pitt beats Louisville by playing well Pitt's NET rating will improve. Depending on what the teams around them do they could go up or down or stay the same. But if you play well your NET rating goes up.
Yea. But I think he is saying he isnt expecting a 25 point win, which is what we need to "break even."
It won't take anywhere near a 25 point win to "break even". In fact it might not take as many as 10.
A 25 point win would most likely be (relatively speaking) a nice boost to our NET.
Like Joe said, any win over around 10 will move us up.
25 would be if we won at home. You may be right with 10 on the road. That still seems low though.
In reality, we won by 13. Our NET ranking before the game was 52. Our NET ranking this morning is, ah, still 52.
Our Pomeroy rating went up slightly, but our ranking also stayed exactly the same, 53. In Torvik our rating actually went down slightly, but our ranking went from 49 to 48.
In short, I think we just about found the break even point for yesterday's game. And it happened to be pretty much exactly where the game ended up.
NET Anomalies
Team B has a better overall record, better, road, neutral and home record, better Q1 record , Better Q2 record
Only difference in favor of team A is number of Q3 wins vs Q4 wins.
Very interesting that who you play seems more important that who you beat. Pitt needs to find out how to minimize Q4 opponents.
BTW Team A is Alabama and Team B is Uconn
Team B has a better overall record, better, road, neutral and home record, better Q1 record , Better Q2 record
Only difference in favor of team A is number of Q3 wins vs Q4 wins.
Very interesting that who you play seems more important that who you beat. Pitt needs to find out how to minimize Q4 opponents.
| NET Ranking | Overall | Road | Neutral | Home | Q1 | Q2 | Q3 | Q4 | |
8 | Team A | 13-6 | 2-2 | 2-3 | 9-1 | 4-4 | 1-2 | 4-0 | 4-0 |
9 | Team B | 17-2 | 3-2 | 4-0 | 10-0 | 7-2 | 2-0 | 0-0 | 8-0 |
BTW Team A is Alabama and Team B is Uconn
The Q4 games helped us I think because we blew them out but those games come with a lot of risk because if you sleepwalk through a 14 point win, it will really hurt you. My thing is you need to play Q4 games, play them on the road
