PDA

View Full Version : NBA Draft Lottery History from 1994-2009 versus the odds



Hicks
01-07-2010, 09:43 PM
2011 UPDATE:

Upon review, my numbers are not accurate. I had falsely assumed the odds remain the same each year (at least starting with 1994), I inadvertently produced the wrong odds in my thread.

I had assumed that team records and tie breakers first determined the order of the draft, and then once the order was established, I assumed that the number of lottery balls was always the same for each slot. I was wrong. They are NOT the same. Turns out, if teams tie in record, FIRST they SPLIT the AVERAGE of the number of balls between both of their slots, and THEN there's a tie breaker to determine who gets the extra ball (if applicable). Looking again at draft express, the ball count for teams changes every year. Even the #1 slot may not receive the usual 250 balls. If that team ties with another team, they SPLIT the number of balls the 1st and 2nd slot usually receive.

So if you are the #1 slot in the lottery, you MAY NOT NECESSARILY HAVE 250 balls

===============

For anyone wondering where we stand in the 2010 lottery, I was curious so I decided to take a look at the modern history results of the lotto.

1994 was the first year for the lottery's current system that will still be in use for the 2010 selections.

I wanted to look at the history of the lottery since '94 to compare what was supposed to happen (so to speak) versus what did happen.

Let's look at the #1 overall pick.

1994 - Went to the 4th worst record
1995 - Went to the 5th worst record
1996 - Went to the 2nd worst record
1997 - Went to the 3rd worst record
1998 - Went to the 3rd worst record
1999 - Went to the 3rd worst record
2000 - Went to the 7th worst record
2001 - Went to the 3rd worst record
2002 - Went to the 5th worst record
2003 - Went to the 1st worst record
2004 - Went to the 1st worst record
2005 - Went to the 6th worst record
2006 - Went to the 5th worst record
2007 - Went to the 6th worst record
2008 - Went to the 9th worst record
2009 - Went to the 3rd worst record

In 16 years, the worst team has won the lottery twice (12.5%)

The 2nd worst team has won the lottery once (6.25%)

The 3rd worst team has won the lottery five times (31.25%)

The 4th worst team has won the lottery once (6.25%)

The 5th worst team has won the lottery three times (18.75%)

The 6th worst team has won the lottery twice (12.5%)

The 7th worst team has won the lottery once (6.25%)

The 8th worst team has never won the lottery.

The 9th worst team has won the lottery once (6.25%)

The 10th or worse team has never won the lottery.

================================================

Now let's look at the #2 overall pick.

1994 - Went to the 1st worst record
1995 - Went to the 1st worst record
1996 - Went to the 3rd worst record
1997 - Went to the 5th worst record
1998 - Went to the 5th worst record
1999 - Went to the 1st worst record
2000 - Went to the 4th worst record
2001 - Went to the 8th worst record
2002 - Went to the 2nd worst record
2003 - Went to the 6th worst record
2004 - Went to the 4th worst record
2005 - Went to the 1st worst record
2006 - Went to the 2nd worst record
2007 - Went to the 5th worst record
2008 - Went to the 1st worst record
2009 - Went to the 6th worst record

In 16 years, the worst team has won the 2nd pick five times (31.25%)

The 2nd worst team has won the 2nd pick twice (12.5%)

The 3rd worst team has won the 2nd pick once (6.25%)

The 4th worst team has won the 2nd pick twice (12.5%)

The 5th worst team has won the 2nd pick three times (18.75%)

The 6th worst team has won the 2nd pick twice (12.5%)

The 7th worst team has never won the 2nd pick.

The 8th worst team has won the 2nd pick once (6.25%)

The 9th or worst team has never won the 2nd pick.

================================================

Finally, let's look at the #3 overall pick.

1994 - Went to the 2nd worst record
1995 - Went to the 4th worst record
1996 - Went to the 1st worst record
1997 - Went to the 2nd worst record
1998 - Went to the 1st worst record
1999 - Went to the 13th worst record
2000 - Went to the 1st worst record
2001 - Went to the 5th worst record
2002 - Went to the 1st worst record
2003 - Went to the 2nd worst record
2004 - Went to the 2nd worst record
2005 - Went to the 5th worst record
2006 - Went to the 3rd worst record
2007 - Went to the 4th worst record
2008 - Went to the 3rd worst record
2009 - Went to the 4th worst record

In 16 years, the worst team has won the 3rd pick four times (25%)

The 2nd worst team has won the 3rd pick four times (25%)

The 3rd worst team has won the 3rd pick twice (12.5%)

The 4th worst team has won the 3rd pick three times (18.75%)

The 5th worst team has won the 3rd pick twice (12.5%)

The 6th, 7th, 8th, 9th, 10th, 11th, and 12th worst teams have never won the 3rd pick.

The 13th worst team has won the 3rd pick once (6.25%)


================================================

Math versus History thus far:

Mathematically, the 1st, 2nd, and 3rd pick are most likely to go to the 1st, 2nd, and 3rd worst teams (obviously).

Historically, it has yet to ever go exactly that way for all three. Only once have the 3 worst teams all won a top-3 pick at the same time (in 1996, 1st got #3, 2nd got #1, and 3rd got #2).

Since 1994, (I rounded up/down when necessary, but I'm listing the actual odds of each, so they won't match up perfectly):

The overall worst team "should have won" the lottery four times (25%), but it has in fact won it only twice (12.25%).

It "should have won" the #2 pick three times (21.5%), but it has in fact won it five times (31.25%).

It "should have won" the #3 pick three times (17.8%), but it has in fact won it four times (25%).

Therefore, historically speaking, if you have the worst record, you're more likely to end up with the 2nd pick, then the 3rd pick, and then the 1st pick in that order.


The 2nd worst team "should have won" the lottery three times (19.9%), but it has in fact won it only once (6.25%).

It "should have won" the #2 pick three times (18.8%), but it has in fact won it only twice (12.5%).

It "should have won" the #3 pick three times (17.1%), but it has in fact won it four times (25%).

Therefore, historically speaking, if you have the 2nd worst record, you're more likely to end up with the 3rd pick, then the 2nd pick, then the 1st pick in that order.


The 3rd worst team "should have won" the lottery twice (15.6%), but it has in fact won it five times (31.25%).

It "should have won" the #2 pick three times (15.7%), but it has in fact won it only once (6.25%).

It "should have won" the #3 pick twice (15.6%), and it has in fact won it twice (12.5%).

Therefore, historically speaking, if you have the 3rd worst record, you're more likely to end up with the 1st pick, then the 3rd pick, then the 2nd pick in that order.

As of this post (1/7/2010), the Indiana Pacers have the 5th worst record in the NBA.

Mathematically, their odds of winning the 1st, 2nd, or 3rd pick in the lottery are 8.8%, 9.7%, and 10.7% -- 29.2% for winning something (1, 2, or 3)

Historically, they're looking at 18.75%, 18.75%, and 12.5% -- 50% for winning something (1, 2, or 3)

Food for thought.

http://en.wikipedia.org/wiki/NBA_Draft_Lottery
http://www.draftexpress.com/nba-draft-lottery-history/
http://www.draftexpress.com/nba-draft-lottery-history/winners/

imawhat
01-07-2010, 09:51 PM
Great stuff that I'd been pondering. Thanks for posting!

If anything this should mathematically and historically eliminate all suggestions of tanking. You don't tank for a 5-10% margin of luck (3rd worst to worst record), especially when the max chance for a top pick is 25%.

MillerTime
01-07-2010, 09:56 PM
nice post....Someone like John Wall alongside Granger and Hibbert could really turn this franchise around

Pacerized
01-07-2010, 09:58 PM
What position was Orlando in when they drew the #1 pick for Shaq?

Shade
01-07-2010, 09:58 PM
So, the 3rd worst record would be ideal. We definitely don't want to finish below 6th.

wintermute
01-07-2010, 10:02 PM
the problem is that the sample size is too small - only 16 lotteries to date. this is what probably produces the mismatch in the "mathematical" and "historical" odds. in the long term, the mathematical odds should hold. still, it's interesting to know how past drafts played out.

Shade
01-07-2010, 10:03 PM
What position was Orlando in when they drew the #1 pick for Shaq?

Orlando had the second-worst record behind Minnesota.

the jaddler
01-07-2010, 10:04 PM
well thank you for that info.....and im not sure if this is the place for it....but lets just throw the draft lottery out!!!!

Shade
01-07-2010, 10:05 PM
well thank you for that info.....and im not sure if this is the place for it....but lets just throw the draft lottery out!!!!

Problem is, that leads to tanking.

Naptown_Seth
01-07-2010, 10:35 PM
A Priori called and it's pi**ed.

;)



Each lottery is a new lottery with no history of the past. Stern, on the other hand, may or may not be so mutually exclusive in determining the probability of winning the lottery. :D

And the reason you don't want to tank is the 25% chance of winning the lottery. Not worth the risk to be that bad. 75% chance you come up short.

Will Galen
01-07-2010, 11:40 PM
Actually a top five pick should give us someone we could really use.

And you don't actually have to get the top pick to win the lottery. The consensus number one doesn't aways turn out to be the best player. For instant MJ was a number three pick.

I want Wall, but I know that's far fetched, so I would be happy to get someone to go with Granger and Hibbert, and the chances of doing that are better the higher in the draft you pick.

CableKC
01-08-2010, 01:57 AM
Can someone tell me if the next draft is going to be like last seasons where there a few top picks and then drops off from there?

the jaddler
01-08-2010, 02:21 AM
Problem is, that leads to tanking.

yeah but if you tank for the sole reason of getting the number 1 pick....you wont be able to do anything with that pick........i mean i dont know what the true solution is here.....i dont think you can get a whole team of players to tank.....can you??????

ok does anyone have the press release's or any storys from and about the reasoning for the lottery?:dance:

Will Galen
01-08-2010, 02:58 AM
yeah but if you tank for the sole reason of getting the number 1 pick....you wont be able to do anything with that pick........i mean i dont know what the true solution is here.....i dont think you can get a whole team of players to tank.....can you??????

ok does anyone have the press release's or any storys from and about the reasoning for the lottery?:dance:

All O'B has to do is play the players he's been playing. The losses will bring the team spirit down, which would get the same result as tanking.

I didn't understand your last question. If you want to read about the upcoming draft see this thread.

http://www.pacersdigest.com/showthread.php?t=48913

rexnom
01-08-2010, 04:02 AM
A Priori called and it's pi**ed.

;)



Each lottery is a new lottery with no history of the past. Stern, on the other hand, may or may not be so mutually exclusive in determining the probability of winning the lottery. :D

And the reason you don't want to tank is the 25% chance of winning the lottery. Not worth the risk to be that bad. 75% chance you come up short.
Thank you for your statistical sense post.

GermanPacersFan
01-08-2010, 05:34 AM
Looking at how things work out in Lottery land, we should be clear that the worst record guarantees you everything but the #1 pick, as long as you aren't the Orlando Magic.

Though, with the way the Nets are "playing basketball" this year, i doubt we'll be able to beat them for the worst record, even if we would trade Danny.

So we should have a good shot at a top 3 pick this year, as long as the team doesn't goes onto one of it's streaks that won't get us to the playoffs, but somewhere around the #10 pick.

count55
01-08-2010, 07:50 AM
Rosencrantz And Guildenstern Are Dead Script



Heads.



Heads.



Heads.



Heads.



Heads.



Heads.



Heads.



Heads.



Heads.



Heads.



Heads.



Heads.



Heads.



Heads.



Heads.



Heads.



Heads.



Heads.



Heads.



Bet?



Heads I win.



Again...



Heads.



Heads.



Heads.



Heads.



Heads.



Whoops!



It must be indicative of something
besides the redistribution of wealth.



Heads.



A weaker man might be moved
to re-examine his faith,



for nothing else at least
in the law of probability...



Heads.



Consider.



One,



probability is a factor which
operates within natural forces.



Two, probability is not
operating as a factor.



Three, we are now held within um...



sub or supernatural forces.



Discuss!
What?



Look at it this way.
If six monkeys...



If six monkeys...



The law of averages,
if I have got this right means...



that if six monkeys were thrown
up in the air long enough...



they would land on their tails
about as often as they would
land on their...



Heads, getting a bit of
a bore, isn't it?



A bore?
Well...



What about the suspense?



What suspense?



It must be the law
of diminishing returns.



I still spell about to be broken.



Well, it was an even chance.



Seventy eight in a row.
A new record, I imagine.



Is that what you imagine?
A new record?



No questions?
Not a flicker of doubt?



I could be wrong.



No fear?
Fear?



Fear!



Seventy nine.



I think I have it.
Time has stopped dead.



The single experience of one coin
being spun once has been repeated.



A hundred and fifty six times.



On the whole, doubtful.



Or, a spectacular vindication
of the principle.



That each individual
coin spun individually is...



as likely to come down
heads as tails



and therefore should cause no
surprise each individual time it does.


Heads.

Unclebuck
01-08-2010, 10:24 AM
Having the worst record still gives you the best chance of getting the number 1 pick. History means nothing when it comes to this sort of thing. It is like if you flip a coin 5 times and each time it comes up heads (assuming the coin is not defective) and then you toss it a 6th time, there is a 50% chance that it will be heads and a 50% chance it will be tails. So suggesting that we should pick heads because it is on a roll, or saying the odds are it will be tails - both arguments have no value.

Hicks - one thing that I believe needs to be factored in is they have I think 2 or 3 times over the years changed the weighting system to give the worse teams a better chance at the number 1 pick . When the lotterry started it was a pure blind draw involving 7 teams. I think they have changed the weighting since 1994, although I'm not sure about that

Hicks
01-08-2010, 12:27 PM
A Priori called and it's pi**ed.

;)



Each lottery is a new lottery with no history of the past. Stern, on the other hand, may or may not be so mutually exclusive in determining the probability of winning the lottery. :D

And the reason you don't want to tank is the 25% chance of winning the lottery. Not worth the risk to be that bad. 75% chance you come up short.


Thank you for your statistical sense post.

Condescension FTW!

I thought it was obvious and went without saying that the past doesn't change the odds any other year. I just thought it was interesting to look at and fun to calculate the "historical odds".

Putnam
01-08-2010, 12:45 PM
It is not about sports, but the following book backs up Hicks' argument. Taleb is the guru of Randomness and Uncertainty.

Seth and Rex and Count55 are right, too. Eventually a very large number of trials (which the NBA draft hasn't seen) will show that the worst team is the most likely to win the first pick. But the odds of the worst team winning the first pick in any given year is very small.

EDIT: Note that, after an infinite number of trials, the worst team will still be getting the top pick only 25% of the time.


http://ecx.images-amazon.com/images/I/41nirFmqcwL._SL500_.jpg

Putnam
01-08-2010, 12:50 PM
nice post....Someone like John Wall alongside Granger and Hibbert could really turn this franchise around


Don't you get it?

We aren't talking about John Wall. We are talking about 12.5% of him. What's that worth?

MillerTime
01-08-2010, 12:55 PM
Don't you get it?

We aren't talking about John Wall. We are talking about 12.5% of him. What's that worth?

Seems like we're talking about odds to get the #1 pick with respect to the worst records (of a team) of a given year.

What does 12.5% of John Wall mean?

count55
01-08-2010, 12:55 PM
Wasn't trying to make a point, per se.

Just always loved that scene.

Naptown_Seth
01-08-2010, 01:12 PM
Condescension FTW!

I thought it was obvious and went without saying that the past doesn't change the odds any other year. I just thought it was interesting to look at and fun to calculate the "historical odds".
Oh, my tone wasn't meant that way. It's more that you could see where the discussion might head after that jump off point.


BTW, not just in the lottery but in all your standard gambling and gaming, it's in the interest of the house to keep it random. If for some reason the #5 team card actually did come up statistically more often (like a roulette wheel with a balance issue), teams would take advantage of that. Instead of tanking you'd have some insane battle of tanks and wins to hold the #5 spot instead.

Admittedly that could be pretty entertaining. :D



And I'm in no way knocking the leg work you did Hicks. I did read your post and was interested in seeing the statistical history. It does add context to the situation.

vnzla81
01-08-2010, 01:19 PM
I'll be ok with at least a top five pick.;)

Naptown_Seth
01-08-2010, 01:21 PM
Not only all this, but on top of your chances of getting any specific pick you also have the chances of any one of those draft slots actually producing a quality player. This is not truly random as it involves human choices, but time has shown that GMs are not perfect and predicting the proper order that players should be drafted, going from most talented career to least (injuries included), has proven well beyond anyone's capability.

In other words, not only are your odds short on getting the #1, they are even worse that you will both get the #1 pick AND that player will be the best player in the long term. Kobe, Jordan, Durant...

You don't only not need to be the worst team to do the best in the draft, you don't even need to win the best pick in the lottery in many years.


Barring a great turnaround the Pacers are going to have a shot at getting a pretty big impact player. But they'll still carry strong odds of failing to get one too.

Putnam
01-08-2010, 01:38 PM
What does 12.5% of John Wall mean?

It means that getting into the lottery, and even with the worst record, provides a team with a very unfavorable chance at winning the top pick.


.

Bball
01-08-2010, 01:59 PM
So if getting into the lottery as one of the worst teams looking for a top pick isn't a decent consolation prize for an awful season.... Is being the 9th seed in an 8 seed playoff picture a better place to be? ...Or even being the 8th seed in a weak EC?

I think you have to roll a lost season into a high draft pick or else you just waste that season for all intents and purposes. In the short term it might suck but it's starting to look like we're delaying the inevitable and creating long term ramifications that are more far-reaching than just biting the bullet on a season that's going nowhere anyway.

Bball
01-08-2010, 02:03 PM
...And isn't this the old conundrum that when playing the state lotto your odds are awful but if you don't play at all then they are even worse?

Shade
01-08-2010, 02:10 PM
A Priori called and it's pi**ed.

;)



Each lottery is a new lottery with no history of the past. Stern, on the other hand, may or may not be so mutually exclusive in determining the probability of winning the lottery. :D

True, but the same system of determination is implemented each time.


And the reason you don't want to tank is the 25% chance of winning the lottery. Not worth the risk to be that bad. 75% chance you come up short.

If you're not good enough to make the playoffs, why should it matter if you have the worst or 14th worst record? May as well increase your chances at landing a superstar.

vnzla81
01-08-2010, 03:01 PM
for those who want a high draft pick this year(including me), don't worry Murphy is starting today.;)

d_c
01-08-2010, 06:46 PM
I don't praise tanking, but the way the league works is the way it works.

In this league, you're rewarded for being either really bad or really good. There are many examples, and the Celtics almost singlehandedly proved that over the past 3 years. There is simply no reward for being a 30 something win team.

I agree with anti-tankers that it's morally a lot more correct to play hard and win as many games as you can, but morals aren't what get you anywhere important in the NBA.

The mechanisms of winning and building a winning team in the NBA are nothing like those of the MLB or the NFL. I don't like the way it works in the NBA, but it is what it is.

Hicks
05-19-2011, 10:48 AM
Upon review, my numbers are not accurate. I had falsely assumed the odds remain the same each year (at least starting with 1994), I inadvertently produced the wrong odds in my thread.

I had assumed that team records and tie breakers first determined the order of the draft, and then once the order was established, I assumed that the number of lottery balls was always the same for each slot. I was wrong. They are NOT the same. Turns out, if teams tie in record, FIRST they SPLIT the AVERAGE of the number of balls between both of their slots, and THEN there's a tie breaker to determine who gets the extra ball (if applicable). Looking again at draft express, the ball count for teams changes every year. Even the #1 slot may not receive the usual 250 balls. If that team ties with another team, they SPLIT the number of balls the 1st and 2nd slot usually receive.

So if you are the #1 slot in the lottery, you MAY NOT NECESSARILY HAVE 250 balls

Brad8888
05-19-2011, 11:26 AM
So if you are the #1 slot in the lottery, you MAY NOT NECESSARILY HAVE 250 balls

If you have 125 fully intact males...:D

naptownmenace
05-19-2011, 12:17 PM
Upon review, my numbers are not accurate. I had falsely assumed the odds remain the same each year (at least starting with 1994), I inadvertently produced the wrong odds in my thread.


So are the rest of the numbers correct? Meaning that the team with the worst record has only won the lottery 3 times in the last 16 years?

Hicks
05-19-2011, 12:51 PM
The numbers regarding who won are still fine. But the %'s for winning a specific spot are partially wrong because of the changing amount of lottery balls each time can get year to year based on tied records.