for all those who advocate tanking

thedofd

thedofd

Bench
#1
#1
There's an interesting piece by John Hollinger on ESPN.com (Insider) about what it takes to really tank to improve your lottery odds, what the odds are that your top pick will be a superstar, etc. I don't know what the rules are about posting complete Insider articles; perhaps a moderator can give me some insight. Here's the link if anyone else is an Insider subscriber:

http://proxy.espn.go.com/espn/print?id=2747955&type=story

SPOILER ALERT :eek:
The last line of the column reads:

"So feel free to root for your team to win now -- and worry about the lottery later."
 
Warhawk

Warhawk

Give blood and save a life!
Staff member
#3
#3
fnordius said:
I think that you're free to post excerpts and/or summaries, just not the whole article.
Click to expand...
Correct. If anything small applies to the Kings you can post it. Otherwise, summaries or snippets are allowed. Copying the entire article is not.
 
nbrans

nbrans

All-Star
#4
#4
Well, not so fast. Hollinger didn't advocate tanking for the three teams (Memphis, Boston, Phildelphia) who are already likely to lose more than 50 games - when you're already losing a whole bunch of games, losing more games on top of that is pretty much pointless. He did say that it is worthwhile to lose at least 50 games for purposes of the lottery, meaning the Kings are right on the border. So for the Kings, tanking falls into the worthwhile category.

Here's what Hollinger says:

The logic gets even more tortured for the worst of the worst, however, as they must lose at increasingly prodigious rates to change their fate much. A team that already is headed for 60 losses, after an 11-30 start, would have to become one of the worst teams of all time, going 6-35 in the second half, just to get a five-game drop in the standings.

The sweet spot

That said, if you were going to pursue such a strategy, there's definitely a "sweet spot" between 53 and 62 losses where it makes the most sense.

As the chart shows, tanking for the lottery probably isn't worth the bother if the team is headed for only 45 or 50 losses. But if it's going to lose between 55 and 60 games anyway, then it starts making a lot more sense. The team can give itself odds as much as 7 percentage points better to land the big prize in this range, with a nearly equivalent jump in the odds of coming in second.
Click to expand...
To lose 57 games (right in the sweet spot), the Kings need to win no more than 25. As in they need to win no more than 8 games the rest of the season. As in the Kings absolutely need to tank.
 
J

JB_kings

Starter
#5
#5
The odds are long if you hope to get a franchise big man like Oden in the lottery. They get even longer if you are out of the lottery or in a lottery with no big man like the once a decade window. I'll take the "tank" for the next few months to have a 1 in 100 shot at Oden.
 
J

JB_kings

Starter
#6
#6
nbrans said:
Well, not so fast. Hollinger didn't advocate tanking for the three teams (Memphis, Boston, Phildelphia) who are already likely to lose more than 50 games - when you're already losing a whole bunch of games, losing more games on top of that is pretty much pointless. He did say that it is worthwhile to lose at least 50 games for purposes of the lottery, meaning the Kings are right on the border. So for the Kings, tanking falls into the worthwhile category.

Here's what Hollinger says:



To lose 57 games (right in the sweet spot), the Kings need to win no more than 25. As in they need to win no more than 8 games the rest of the season. As in the Kings absolutely need to tank.
Click to expand...
They are on track for that now with a healthy group of players.
 
thedofd

thedofd

Bench
#7
#7
Here are some of Hollinger's key points:

CAN YOU WIN BY LOSING?

Since the league expanded to 29 teams (now 30 teams), we've had 11 draft lotteries. The team with the best odds -- in other words, the team with the worst record -- has, on average, lost 65.7 games. (Quick technical note: Throughout this study, records for the 1998-99 lockout year are prorated to 82 games.)

But those teams don't always win the drawing. Lottery winners have fared a bit better -- on average, they lose 59.3 games. ...

You might think of 60 losses as the gold standard in terms of giving a team good lotto odds, but in fact that team would be fourth or worse in the pecking order in seven of the past 11 lotteries.

Only when a team loses 64 games or more do the odds get to 90 percent or better of having one of the league's worst three records, thereby guaranteeing a draft choice no worse than sixth.

If the goal is to assure a top-four pick, which requires the worst record in the league, the hill gets even steeper. Based on recent history, it takes 67 losses to get better than 50-50 odds of having the worst record.

WHAT'S A LOSS WORTH?

Running through the 11-season progression, we find a 61-loss season produces, on average, a 15.6 percent chance of winning the lottery. ...

... Going from 45 losses to 50 improves a team's odds of winning the lottery by only 1.99 percentage points, which probably isn't worth the bother for most teams.

TOUGH TO TANK

... Tanking in order to win the lottery is actually incredibly difficult. Even a five-game drop in the standings produces only about a 6 percentage point better chance of winning the big prize in most cases (and if you're wondering about the consolation prize in the Durant/Oden Sweepstakes, the impact is roughly the same).

And for a variety of reasons, a team that is tanking games probably won't drop by more than five additional games.

Think about it for a second. Any team bothering to lose on purpose is already obviously bad, so such teams are bound for 50 losses no matter what.

Additionally, most teams probably wouldn't implement such a strategy until about the midway point of the season, so they'd have only about 41 games to follow through on the effort. ...

The logic gets even more tortured for the worst of the worst, however, as they must lose at increasingly prodigious rates to change their fate much.

FOOL'S GOLD?

Of course, all this assumes that our lotto winners hit the jackpot by ending up with the top overall pick. But that's not always how it works out. Look at the recent drafts when there was a consensus No. 1 pick by draft day, and compare how that player turned out against the player picked, say, fifth. Sure, the odds are in the top pick's favor, but it's by no means a home run.

Even if a team wins the sweepstakes for a no-brainer superstar, as Cleveland did with LeBron James, it might not get the best player in the draft -- as Miami fans certainly would argue for that draft, in which Dwyane Wade was the fifth pick.

Allen Iverson turned out great in 1996, but was he that much better than No. 5 pick Ray Allen? ...

And sometimes the consensus No. 1 flops -- as top pick Michael Olowokandi did in 1998. The fifth pick that year? Vince Carter. ...

That's not to say you'd rather pick fifth, mind you ... there are some years when you'll end up with Nikoloz Tskitishvili instead of Yao Ming , for instance. It's just that it doesn't always turn out so badly for the so-called "loser" in the lottery, and that's another factor in analyzing the "tanking" strategy. For all the research teams do, the draft remains an inexact science.

And ultimately, that might be the biggest argument against tanking games for draft position. Not only is the increase in odds of winning the lottery fairly marginal -- only about 6 percentage points in the best cases, significantly less in others -- but it all gets back to the question of "What are we winning here?"

Look at the 10 players starting in the NBA All-Star Game -- only four of them were taken among the first two picks in their respective drafts. And if you look at Marc Stein's picks for the reserves, only two of those 14 gentlemen were top-two picks.

In other words, superstars don't emanate from only the very top of the draft. ...
 
T

tjobrien

G-League
#8
#8
thedofd said:
There's an interesting piece by John Hollinger on ESPN.com (Insider) about what it takes to really tank to improve your lottery odds, what the odds are that your top pick will be a superstar, etc. I don't know what the rules are about posting complete Insider articles; perhaps a moderator can give me some insight. Here's the link if anyone else is an Insider subscriber:

http://proxy.espn.go.com/espn/print?id=2747955&type=story

SPOILER ALERT :eek:
The last line of the column reads:

"So feel free to root for your team to win now -- and worry about the lottery later."
Click to expand...

I had just gotten done reading this and came here to see what everyone thinks.

I think it's the worst artilcle Hollinger has written in some time. It seems he started with a conclusion, found his data, and then twisted his argument to make the two fit. At one point he says something along the lines of, "The difference between 45 losses and 50 losses is only a 2% chance of getting the #1 pick, and is not worth the trouble." Ignoring the fact that HE ignores that the top THREE picks are all highly coveted this year, let's think about this.

Excuse me, John? 2% isn't worth the trouble? What trouble? What possible good can come from losing 45 games instead of 50? When you are in this situation, every percent chance counts. 5 freaking games. Hell, for a team as bad as the Kings it is more trouble to WIN 5 games than to lose them! I don't understand his logic at all.

Yes, the lottery is a gamble. Literally. But when you are gambling, don't you want every little edge? If you're playing poker and you know that a card could give you a 2% greater chance of winning, don't you jump on that?

I just don't get his logic.
 
thedofd

thedofd

Bench
#9
#9
tjobrien said:
II think it's the worst artilcle Hollinger has written in some time. It seems he started with a conclusion, found his data, and then twisted his argument to make the two fit. At one point he says something along the lines of, "The difference between 45 losses and 50 losses is only a 2% chance of getting the #1 pick, and is not worth the trouble." Ignoring the fact that HE ignores that the top THREE picks are all highly coveted this year, let's think about this.
Click to expand...
Did you read the companion article comparing Durant, Oden and Noah? The conclusion seemed to be that there's a pretty fair drop-off between the first two picks and the third.
 
F

fnordius

All-Star
#10
#10
The difference between 50 losses and 45 losses is a 0.0% increased chance of a western team making the playoffs. There is absolutely no benefit whatsoever to winning those 5 games, so what's his point? That cutting off one's nose to spite one's face is just the right thing to do?
 
T

tjobrien

G-League
#12
#12
fnordius said:
The difference between 50 losses and 45 losses is a 0.0% increased chance of a western team making the playoffs. There is absolutely no benefit whatsoever to winning those 5 games, so what's his point? That cutting off one's nose to spite one's face is just the right thing to do?
Click to expand...

Exactly what I was trying to say. Thank you. It makes no sense NOT to go from 45 losses to 50 losses.
 
J

JB_kings

Starter
#13
#13
First of all, the Kings aren't tanking - they are THAT bad. Call it underacheiving, bad coaching, poor chemistry, whatever. They are right where they belong. It's a poorly written article because it only highlights situations that make his point. It ignores guys like Shaq, Tim Duncan and LeBron that were franchise impact players with the #1 pick. This year there's a good chance that the top 2 players in the draft are franchise guys. Lately the free agent market has been yielding overpriced so-so players and trades for franchise players are more rare than #1 picks that are stars in the league.
 
Krunker

Krunker

Northernmost Kings Fan
#16
#16
tjobrien said:
I think it's the worst artilcle Hollinger has written in some time. It seems he started with a conclusion, found his data, and then twisted his argument to make the two fit.
Click to expand...
You're right on the money. With the worst team in the league is guaranteed to get one of the top four picks, its a big deal when there are some definite standouts at the top (e.g. Oden, Noah, Durant, Wright) and you know you'll get someone good, as opposed to a crapshoot at #10.

The second worst team has an 88% chance of picks < 5, and the third worst team has about a 95% chance of picking < 6, so it really depends on the depth of the draft and where the dropoff point is.
 
Bricklayer

Bricklayer

Don't Make Me Use The Bat
#17
#17
Never a fan of Hollinger, and call him out repeatedly. But here he actually manages to do the double dip of simultaeously missing the point of the tanking argument and twisting numebrs to support his position. Also of course engages in the fallacy of trying to introduce general statistics into a concrete situation.
 
F

fnordius

All-Star
#18
#18
The majority are road games, 21, vs. 18 home games.

And 5 or 6 games against teams with weaker records, vs. 33 or 34 against teams with better records.

Squeaking into the playoffs means something like winning 100% of our remaining home games, and 1/3 or more of our road games -- not gonna happen.

Finishing with a .35 record -- sufficient to get 4th place in the lottery -- would require nothing more than going 12-27 during the remainder of the season -- a definite possibility, given that half of those wins would have to be against stronger teams. If we fail to beat teams that have better records, 3rd place is not out of reach.
 
You must log in or register to reply here.