Discussion in 'Pistons and NBA' started by roscoe36, Aug 29, 2006.
Gooooood Morning Pistons Fans!!!!!!!!!
Hummmm.........I think I remember that guy. Maybe he will show up again someday?
RealGM: NBA Lottery Simulator
lottery simulator. Orlando falling to 14th pick.
The lower the pick falls the better the odds of getting Acie Law.
Well, thats the guy I want so far.
So, my understanding is that we would get 5 balls in the machine if Orlando's current position holds. So, the odds of us getting the #1 pick (which would then go to Orlando) would be 0.5%.
My question is, what are the odds of the pick landing in the top 5? In other words, how likely is it that Orlando will get to end up keeping this pick?
My calculations show that they will have a 6.6% chance of landing a top 5 pick. 1 - [(995/1000)x(745/750)x(544/551)x(388/395)x(269/276)] = .06579...
So, if Orlando finishes 16th in the NBA, they have a 0% chance of getting their pick back (because they wouldn't get any ping-pong balls), but if they finish 17th or worse, they have at least a 6.6% chance of getting a spectacular pick.
This is strange. It is very unlikely that they will get their pick back, but if they do, it will be a team changing pick.
The best Detroit can do is get really lucky and have the Orlando ball picked 6th. In other words, just because we would have the least balls in the lottery does not mean that we will get the 14th pick. Could be the 10th, etc. 11-14 are all very similar odds-wise with 8, 7, 6, and 5 balls in the lottery respectively out of 1000.
How are you doing that calculation, Tashawn? I thought it was only the first three picks that were determined by drawing ping-pong balls out of the popcorn machine.
Should you be multiplying or adding?
I'm having flashbacks to those awful probability class exercises that always seemed to begin with a bunch of black and a bunch of white socks in a box.
There are only 14 balls in the machine. they draw 4 balls and you'll get a combination of 4 numbers:
memphis is given a sheet of 250 possible 4 number combinations giving them the best chance to match those 4 numbers.
Orlando is given 5 sets of numbers to match the winning combination.
The balls all go back in and the process is started over again.
Orlando has .5% chance to get any of the top 3 picks. After that, its based on their record.
Forgot about the top three and then based on record... This definitely requires a computer.
Yes. You need it to google.
I just took the 1 minus the odds of NOT getting a top 5 pick. So each of those fractions would be multiplied together. The worst team gets 250 balls in the machine, so I subtracted 250 from the numerator and denominator after the first pick and did the same for each one after that, using that team's number of balls. Based on Slippy's comment, they quit drawing after the 3rd pick? I didn't know that, but that would explain why the on-line simulators always put us in 14th place.
I think in the past (1980's), only the top 3 would get a shot at it, but they changed the rule to allow 14 teams in the 90's. They later changed it again to weight it more heavily for the worse teams when Orlando won the lottery 2 years in a row as the underdog. The second time they won it, they only had a 1 in 66 chance of it.
Remember when Milwaukee got the top pick a couple years ago? They had the 6th worst record in the NBA. So their odds of getting the top pick were only 63/1,000 = 6.3%. They picked Bogut.
The Magic would have a 1.81% chance of getting a top 3 pick I believe. (.5% of getting #1, slightly better of getting #2, and even better of getting #3).
This does not affect anything, but in reality, there are not 1,000 balls in the lottery machine. There are only 14. They draw 4 balls at a time and that will yield 1 of 1,001 different combinations (3, 12, 23, 5 for example). Each team is assigned a # of combinations. So, the Pistons would get 5 of these combinations assigned to us, while Memphis would get 250 of these combos.
The reason that this high number of balls only yields 1,001 combinations is that the order of them does not matter. So the above example would be expressed as (3, 5, 12, 23). It's more like playing BINGO than having your ball drawn. Notice that there are 1,001 combos and there should only be 1,000. So, there is one combo that they don't give to any team. That would be ignored if it were ever picked. It is something like 11, 12, 13, 14. (again, what Slippy said... it just took me a while to write this post).
yeah...lottery is a misnomer. Its more like Bingo...or as this year, it could be ODOM.
also, 11-12-13-14 is the black sheep combo. for this i think it means the #1 prospect has to stay in school.
how much better are the chances for the 2nd and 3rd picks?
if memphis wins the #1 spot, you scratch off 250 possible combinations.
at that point Orlando gets .66% ??
then if boston gets #2, thats another 199 combinations. so thats .9%??
I don't know if my math is correct.
Also, I just ran the realgm thing and orlando's pick moved to 13. ACIE!!!
You can see the process described and the odds calculated here, at Wikipedia. For the purposes of our discussion: if the Magic finish the season with the best record out of all the lottery teams, i.e. the 14th worst record in the league, that means the percent chance that they get the 14th pick (which would then go to the Pistons) is 98.2%.
So the pick is top 5 protected? In effect, that means:
(a) If the Magic finish with one of the five worst records in the league, they keep their pick no matter what.
(b) If they finish with one of the 6th through 14th worst records in the league, the only way they keep their pick is if they draw one of the combinations that gets them into the top 3 picks of the draft.
By the way, calculating the chance that any particular team gets the #2 pick is a bit messy, though not too bad. But calculating the chance that they'll get the #3 pick? Oy vey...not pretty. You'd definitely need to write a program to calculate those numbers (fortunately, they're already right there in that Wikipedia link).
Looking at these numbers, I've even more confused. The team with the worst record always seems to get a 25% chance at getting the first pick, but then the odds for the other teams appear to be based somehow on their records, not just on the order in which they finished. The odds change each season, but I don't understand how they're calculating it.
Here is a good experiment. Try refreshing the simulator over and over until the Orlando pick jumps from 13 to the top 3. According to my math, it should happen on average every 60 resets.
The first time I tried it, it took 44 resets... and the Orlando pick was #1. Ouch.
The Realgame sim is smart enough that it changed the Detroit logo to Orlando when they won it... b/c it knew that we have to give the pick back to them.
39 resets and they got #1 again.
Now, this outcome should average 1 in 200, so this is an abberation. It could easily take 300 resets before seeing it again.
I think I finally get it. Slips was right with the top-3 are picked in the bingo system only. After that, everyone gets sorted out according to record.
In conclusion, and after a little more research, I can conclusively say that the odds of us losing our pick to Orlando are the following for each of the final finishing positions of their team:
14th pick- 1.8% (chance of losing it to Orlando)
13th pick- 2.2%
12th pick- 2.5%
11th pick- 2.9%
10th pick- 4.0%
9th pick- 7.9%
8th pick- 8.2%
7th pick- 18.3%
6th pick- 18.3%
5th pick- 100.0%
NBA Draft Lottery - Wikipedia, the free encyclopedia
These percentages are taken by adding up the odds for 1st, 2nd, and 3rd for each spot from the table.
Edit- now I see that Dumars4Ever said most of this too.
Tashawn, you were probably putting that post together without seeing the few I did above it...strangely, the odds seem to change a bit each year based on the teams' records, not just on the order in which they finish. I don't quite get how they calculate those numbers.
Oh, I think I know why. The year that was posted on Wikipedia had a few teams that were tied. They probably didn't get as big an advantage over the team that they won the tiebreaker from (coin flip used).
Timberwolves/Celtics for example.
It's a little convoluted, but it makes decent sense once you understand the rules.
Basically, if Orlando tanks a few more spots, it really doesn't do much for them in terms of getting the pick back. What is a 2% chance vs. a 1.8% chance? That is all I was really going after, was to see if they had a tanking incentive late in the year. I would say that no they do not.
And those weird diagonal sets of numbers off on the right hand side of the table. Those just tell you how likely it is that a team outside of the top 3 will crack into the top 3. Because if that happens, then they resort all the remaining 11, and you will get pushed back a spot. If 2 crack in, you'll get pushed back 2, and if 3 do it, then you'll be pushed back 3 (very unlikely).
If you look at the Hawk's odds, it is very strange.
Here are their odds of landing the following spots:
It is really strange that the distribution is not a bell curve. 4th was their 2nd least likely spot and it is surrounded by more likely spots on both sides.
I don't think its feasible for them to drop to the 5 spot. Even if the lose every game, I don't see how far the can really drop.
Separate names with a comma.