Stats: T. Allen > Daniels

[Golabki]

With Daniels finely in the fold I wanted to compare him with some of the other Celtics wings. I looked at adjusted +/-, which essentially measures much better or worse (in points) a team plays with a given player on the court (it is much or rigorous that traditional +/- or on/off court stats, but it's the same idea). If you want to know more go to basketballvalue.com. Anyway, in my opinion this is easily the most complete and least biased stat available. I compared Pierce, Ray, Tony, House and Daniels over the last 5 years.

Pierce + 7.9
R Allen + 5.4
T Allen + 4.9
Daniels - 1.1
House - 1.5

To put these numbers in perspective Pierce and R Allen's numbers put them at all-star level (but clearly below superstars like LeBron). Daniels and House's numbers put them as solid bench player. So, those four guys basically fall in line with general expectations.

Tony Allen is obviously the surprise, being ranked as a all-star (or almost all-star) type player. If you dig a bit deeper the system actually ranks Tony as a pretty mediocre offensive player (which we can all agree is the case), but his value comes from one of the best defensive ratings for any perimeter player.

I have to say the Daniels signing looks pretty good by these numbers (solid bench player for almost no money is good), but I hope Tony is still given a chance to show his value.

[chas0x01]

Tony Allen only played 46 games and I believe most of that was due to injuries. But he's a very good bench player for sure. I was hoping that if he were traded in some package that he'd get boughtout, and we'd re-sign'm for the vet min. It would be great to have him around to help deal with some of the super explosive PG in the league.
What hurts TA is his poor handle and inability to play any kind of effective PG on the offensive end. We would get the most out of House if we had a big guard who could do that, which we do, now - Daniels.
This is actually a pretty sweet team when you really think about it.

[SonicYouth34]

I like TA a lot. But he needs lots of mins to be productive, something he won't get here in Boston. He also has the mental issues and again, doesn't have the mins to work through them.

[GuyClinch]

TA is underrated around here - but his offensive game has been bad enough such that Doc simply will not play him. Doc talks defense but he plays offense.

[sunshinekids99]

Well you do need to actually score to win basketball games. Tony Allen disappears way too much in games.

[Ed Pinkney]

"People can come up with statistics to prove anything. 14 percent of all people know that."


I agree with starters minutes Allen could possibly finally put it together. Could. But when I watch him play, I just don't see a fundamentally good basketball player. Spotty shooter, suspect ball handling and decision making, silly fouls/mistakes etc is what you see far too regularly.

[Jammer]

Adjusted plus/minus sounds nice in theory, but when I investigated that guys website a few years ago his results were not tracking reality. The idea may be good, but there is something wrong with his method, which yields to results, or conclusions, that at the time I felt were just plain wrong.

If you want a rigorous stat, I prefer NET PER, which is your PER, minus your MAN's PER. That's brutal. The Roland Rating is a combination of NET PER (weighting has been going up, was 60%, may now be 75%) and straight on court/off court plus/minus; although at one time the Roland Rating (probably in 2003-2004 season) was a straight ADJUSTED PLUS/MINUS ranking.

[MyInsatiableOne]

You could come up with some formula and stats to prove Greg Kite was better than Kareem, too...

Not buying it...

[chas0x01]

TA can make you nervous everytime he dribbles the ball. And he can make some stupid plays for sure. But overall TA's still one of the best perimeter defenders in the league. And he scores at the same level as Baby and Powe, with a very good FG% for a guard. I'm glad he's still on the team.

[MyInsatiableOne]

TA can make you nervous everytime he dribbles the ball. And he can make some stupid plays for sure. But overall TA's still one of the best perimeter defenders in the league. And he scores at the same level as Baby and Powe, with a very good FG% for the guard. I'm glad he's still on the team.

That makes one of you...

Too bad the Celtics themselves don't agree with you...

[Rocky5000]

It's good to have both TA and Daniels, because both of those guys are more injury prone than the average player. I think Bill Simmons started the TA hate train and then a lot of people jumped on his caboose.

We can't be sure that Daniels will work out any better than TA in our system, but he gives us flexibility in the backcourt and insurance. Both players can't shoot from the outside (TA actually shot slightly better), and score most of their points in the paint, TA is a better defender than Daniels, and Daniels is a better ball handler than Tony.

[sully00]

Adjusted plus/minus sounds nice in theory, but when I investigated that guys website a few years ago his results were not tracking reality. The idea may be good, but there is something wrong with his method, which yields to results, or conclusions, that at the time I felt were just plain wrong.

If you want a rigorous stat, I prefer NET PER, which is your PER, minus your MAN's PER. That's brutal. The Roland Rating is a combination of NET PER (weighting has been going up, was 60%, may now be 75%) and straight on court/off court plus/minus; although at one time the Roland Rating (probably in 2003-2004 season) was a straight ADJUSTED PLUS/MINUS ranking.

Both stats don't tell you a damn thing by themselves. As a matter of fact they would both greatly overrate TA. Not because Tony is good or bad but the people that developed did not develop them as stand alone stats. The Roland Rating while flawed in its ability to reflect reality is at least an attempt at a rating.

The fact that TA has a NETPER of +4.3 does not indicate that he is a better player, or more importantly even a higher rating than Eddie House who has +3.9. To use this ratings you have to start with the actual PER, if you don't crack 15 you aren't rating very well. Eddie House at 17 is rated higher than TA at 14.5 no matter what the differential is.

The idea isn't to rely on a single stat to tell you anything, you need to digest what they all tell you and then weigh it against reality.

As far as adjusted +/- the biggest issue in this instance is that Daniels starter for half his games and came off the bench for half of his games so you would need the splits on that to compare him to TA.

[Celts17Pride]

Eyes: T. Allen sucks < Daniels good role player

[Bluewhale]

Tony Allen is a turnover prone expert. That's the least help we need.

[MyInsatiableOne]

Tony Allen is a turnover prone expert. That's the least help we need.

And that's his *best* quality!

[greenbeans]

Basketball: Daniels > Allen

[Golabki]

Some are saying, "I watch the games, and Tony sucks, he can't dribble and can't shoot".
I agree to an extent. There is no question Tony is a poor ballhandler and shooter for a 1, 2, or 3. That has been pretty obvious since his rookie year. And the stats I mentioned above also agree with this. However, his offensive game is really not that bad for a defensive speciallist, which is what Tony should be.

To the points about the stat itself...
Adjusted plus/minus is the best stat available by a wide margin. The reason for this is very few assumptions go into the number. Here is how the stat works... Team A plays Team B (with the same 10 guys on the court for the whole game) and they tie. Then one player on Team A as taken out and replaced with a new guy. The game is replayed and Team A wins by 5. So, the new guy is 5 points better than the old guy. The assumption is that if your team scores more/gives up less with you on the court, than you are a good player. The math is hard, but the idea is very simple. PER and NetPER, on the other hand both make huge assumptions and omit major parts of the game. PER and NetPER are very similar to just Adding Point, Assists, Rebounds, Steals, Blocks and subtracting TOs, or just eyeballing a guys stat line. Adjusted plus/minus is a fundementally better approach. The flaw with Adjusted plus/minus is that there is a lot of random fluctuation in the numbers, so you need a relatively large body of work to get a number you can have confidence in. This may be why people feel the numbers don't "track reality" if you look at any given year. That's why I took 5 years.

I should also point out it would be an overstatment to look at these numbers and say Tony Allen is as good, or almost as good as Ray Allen. However, I think you can say Tony Allen very likely underrated as a player and better (overall) than Daniels and House.

[MyInsatiableOne]

Life: Daniels > Allen

Fixed

[sully00]

golabki

The stat itself is an assumption. What it calculates never occured. It also trying to attach individual value to a team calculation. On its face it would seem incredibly unfair to compare players on the Pacers with players on the Celtics based on game score. I am not saying. the stat doesn't have merit and is far more useful than raw +/- but these measures were not about simplifying player comparison quite the opposite it is about getting a more in depth or complex look.

When you want to make a point about two players then use as many of these measures as you can find and understand and compare the results. The fact that a stat says the Paul Pierce and Ray Allen play at an all star level is not validating, if didn't it would be disqualifying every assessment should state the obvious.

But to what this thread should be about

Basketball: Daniels > Allen

Anyone who is expecting a significant upgrade from TA in Daniels is in for a rude awakening. While Daniels has shown to be a much more productive player as a starter throughout his career

Marquis Daniels 47% FG 14.5 ppg 5 rpg 3 apg 1.8 topg 35 mpg in 109 starts
Tony Allen 50% FG 10.9 ppg 4 pg 2 apg 1.9 topg 26 mpg in 74 starts

this does not hold in a reserve role. I have always felt that TA was equipped to be a starter that may not be the case. The difference, or lack there of, in production in a reserve role should grab your attention.

Marquis Daniels 45% FG 7.2 ppg 2.5 rpg 1.6 apg 1.17 topg 18 mpg in 242 games
Tony Allen 46% FG 6.2 ppg 2.2 rpg 1.1 apg 1.27 topg 16 mpg in 208 games

Both are considered good defenders, I think TA is more talented in this regard and defensive measures support that, we are hoping that Daniels is little smarter and makes better decisions with the ball in his hands. Both are particularly fragile players.

Anyone coming at this thinking that TA is garbage and Daniels is going to be great and a huge upgrade is talking out their ass. The hope is that Daniels is a better fit as a primary ball handler when paired with House and possibly a better on the ball defender on opponents stars were TA has had a tendency to get silly fouls. But overall we are talking about two guys with similar skill sets, one who has shown to be a capable starter and the other might be a little bit more productive bench player. The one significant upgrade Daniels brings is the ability to step in and start at the SG/SF and give this team capable starter production.

In the end the move is to make Eddie House who has the NBA skill we need to exploit, 3 pt shooting, more effective if Daniels can do that then it is a good move if not then I wouldn't be shocked to see Doc go back to TA.

Splits and situations are so much more important in evaluating what has happened in an effort to predict what will happen. More information is always better than less. What I would really like to see happen with these advanced stats is more splits available, that is difficult with PER in any form because its reliability becomes really inconsistent the smaller the data set. But what I want to see is House's performance with the first unit vs with TA and the same with TA. Who would be better replacing Rondo with the first unit Daniels or House?

I would love to see 82 games and Popcorn Machine kind of merge players individual stint production with 5 unit performance.

[hairybyrd]

I like TA for several reasons but he still has a smaller brain than Daniels. Years ago he was asked what ONE word best exemplified the Boston Celtics and he said "hard work."