Spurs are nearly locked into having the best odds to land the number 1 pick. So I ran 10,000 draft simulations. Here are the results

[Dex]

Looks to be higher than 14% though if my eye is guesstimating this properly. Looks to be closer to 16% or 17%. Not sure I understand the statistics behind that.

Technically when you flip a coin, you have a 50% chance for both heads and tails.

That doesn't mean that out of 100 flips, the final count will be 50 and 50.

[playblair]

is it u?

spurstalk getting plagiarized again.......

[exstatic]

Technically when you flip a coin, you have a 50% chance for both heads and tails.

That doesn't mean that out of 100 flips, the final count will be 50 and 50.

We have a 14 % chance. If you run a simulator 10 times, it's impossible to win 1.4 times. Sample size is everything.

[scott]

Technically when you flip a coin, you have a 50% chance for both heads and tails.

That doesn't mean that out of 100 flips, the final count will be 50 and 50.

But that’s only because a sample size of 100 is too low. As the number of iterations increase, the distribution of heads/tails will converge upon 50/50. It’s a beautiful thing, but it is also what’s leads people to be in dumbfounded shock when they watch a roulette wheel land on red 23 times in a row and the casino collects all the chips they just lost.

[offset formation]

I think offset formation meant the particular 10,000 run and bar graph that PB did, not that the ACTUAL odds aren't 14%.

Yes. Which is why I highlighted the apparent discrepancy between the 14% and what appeared to be a bit higher bar graph.

[offset formation]

do u approve of my new sig

I told you to post whatever you wanted but just be prepared for the feedback if it was demonstrating ignorance, then proceeded to give you reasons why it was that.

[offset formation]

Technically when you flip a coin, you have a 50% chance for both heads and tails.

That doesn't mean that out of 100 flips, the final count will be 50 and 50.

Yes but this was a run of 10000 runs which ought to settled out at the 14% number by then, or at least I would have thought so. Guess it takes a billion though.

[Sugus]

Yes but this was a run of 10000 runs which ought to settled out at the 14% number by then, or at least I would have thought so. Guess it takes a billion though.

Exactly, and it's the reason these types of "ten thousand/million/billion times" probabilistic analysis are dumb in the first place for single-time events.

The draft lottery isn't going to be ran a number of times high enough to validate the percentages. It's up to dumb, random luck whether the Spurs have the ping-pong balls fall their way or not, and having 14% chances, or 12% or 16%, isn't a significant difference when the chances are already so low to begin with.

We either get Wemba or not, I dunno why everyone stresses about the numbers on it. Spurs have done their job, and now it's up to fate itself.

[Phenomanul]

But that’s only because a sample size of 100 is too low. As the number of iterations increase, the distribution of heads/tails will converge upon 50/50. It’s a beautiful thing, but it is also what’s leads people to be in dumbfounded shock when they watch a roulette wheel land on red 23 times in a row and the casino collects all the chips they just lost.

Well... that's because Roulette wheels are rigged (everyone knows that the Roulette operator can bias where the ball will land). On the other hand, try flipping an unrigged coin with a stated objective to get 23 consecutive heads... you would grow old before that happened.

[Phenomanul]

As an aside, I believe the NBA draft lotto can also be rigged.

[exstatic]

As an aside, I believe the NBA draft lotto can also be rigged.

And how would that be done?

[Phenomanul]

And how would that be done?

Any system can be rigged.

[Phenomanul]

If I had the engineering specifications for the machine that was used, the specifications for the ping pong balls, the materials of construction, if we knew all of the relevant operating parameters (such as the flow of air into those machines, etc...) then one could technically (by definition) develop dynamics that would counter the machine's stated design objectives. It's called devolution, reverse engineering... I'm not saying I know the precise way that the 2023 NBA Draft Lotto machine can be rigged, I'm just stating that it can be rigged. That distinction, is the point I am trying to make here.

[scott]

Well... that's because Roulette wheels are rigged (everyone knows that the Roulette operator can bias where the ball will land). On the other hand, try flipping an unrigged coin with a stated objective to get 23 consecutive heads... you would grow old before that happened.

Of course, and 23 is also a random number I made up... but the bigger point was that after you've hit 22 in a row, the odds of the 23rd are still 9/19 in roulette and 50/50 for a coin flip. The results of the previous X number are irrelevant to future outcomes.

Personally, the most consecutive strikes in a row I've seen in person in a casino is 13. Allegedly some American casino hit red 32 times in a row back in the 40s. Odds approximately 1 in 11 million. The odds of 10 consecutive are approximately 1 in 1400. "Slower" roulette games tend to run about about 30 spins per hour, so the expected value is that the typical roulette wheel will hit a 10-in-a-row every two days or so. Not that uncommon at all.

I actually think casino roulette wheels have a lower likelihood of being rigged than the NBA lottery. Casinos already have math on their side, they don't need to rig the roulette wheels.

[MannyIsGod]

If I had the engineering specifications for the machine that was used, the specifications for the ping pong balls, the materials of construction, if we knew all of the relevant operating parameters (such as the flow of air into those machines, etc...) then one could technically (by definition) develop dynamics that would counter the machine's stated design objectives. It's called devolution, reverse engineering... I'm not saying I know the precise way that the 2023 NBA Draft Lotto machine can be rigged, I'm just stating that it can be rigged. That distinction, is the point I am trying to make here.

Ah yes the NBA is precisely engineering ping pong balls in a secret ballot that no one ever sees to come out in a certain way when they could just make up instead. The NBA Draft lotto machine isn't rigged in this manner. This is pretty ridiculous.

[Ariel]

If I had the engineering specifications for the machine that was used, the specifications for the ping pong balls, the materials of construction, if we knew all of the relevant operating parameters (such as the flow of air into those machines, etc...) then one could technically (by definition) develop dynamics that would counter the machine's stated design objectives. It's called devolution, reverse engineering... I'm not saying I know the precise way that the 2023 NBA Draft Lotto machine can be rigged, I'm just stating that it can be rigged. That distinction, is the point I am trying to make here.

I posted the procedure in the previous page:
https://www.nba.com/news/nba-draft-lottery-explainer

Fourteen ping-pong balls numbered 1 through 14 will be placed in a lottery machine. There are 1,001 possible combinations when four balls are drawn out of 14, without regard to their order of selection. Before the lottery, 1,000 of those 1,001 combinations will be assigned to the 14 participating lottery teams. The lottery machine is manufactured by the Smart Play Company, a leading manufacturer of state lottery machines throughout the United States. Smart Play also weighs, measures and certifies the ping-pong balls before the drawing.

The drawing process occurs in the following manner: All 14 balls are placed in the lottery machine and they are mixed for 20 seconds, and then the first ball is removed. The remaining balls are mixed in the lottery machine for another 10 seconds, and then the second ball is drawn. There is a 10-second mix, and then the third ball is drawn. There is a 10-second mix, and then the fourth ball is drawn. The team that has been assigned that combination will receive the No. 1 pick. The same process is repeated with the same ping-pong balls and lottery machine for the second through fourth picks.

If the same team comes up more than once, the result is discarded and another four-ball combination is selected. Also, if the one unassigned combination is drawn, the result is discarded and the balls are drawn again. The length of time the balls are mixed is monitored by a timekeeper who faces away from the machine and signals the machine operator after the appropriate amount of time has elapsed.

So it requires multiple balls, sequence doesn't matter and each team is assigned a number of combinations. If no team contains a disproportionate amount of times a given ball (particularly teams with smaller odds), even if a given number is more likely to appear than others, it shouldn't have such a huge effect on the outcome. Also you could keep it running for a while before the process starts, and if a given number keeps popping up it should be easy to spot. I'm pretty sure NBA franchises aren't oblivious to these things, and fixing the outcome isn't as easy as coming up with the specifications of the machine.

[Ariel]

Yes but this was a run of 10000 runs which ought to settled out at the 14% number by then, or at least I would have thought so. Guess it takes a billion though.

It doesn't, I just took a ridiculously high number of tries because 1) it takes a few seconds and 2) wanted to make a point. But even 10000 tries should converge relatively quickly to within 1% of the true mean:
mean(rand(1,10000) < 0.14)
5 tries: 0.1404, 0.1411, 0.1389, 0.1392, 0.1425
try here: https://ideone.com/l/octave (paste code and run)
If it doesn't it's because the simulation was not properly conducted (no code posted, so) or there was no simulation at all, simply a fixed chart to elicit a response (IMO the case here). My point is, there is no need to simulate anything, because there are 2 possible outcomes and the probabilities of each are known beforehand, so it adds no value (it would if you didn't know the odds and had to estimate them, for instance). The whole concept of this thread is just stupid, IMO the OP is just trolling.

[MannyIsGod]

It doesn't, I just took a ridiculously high number of tries because 1) it takes a few seconds and 2) wanted to make a point. But even 10000 tries should converge relatively quickly to within 1% of the true mean:
mean(rand(1,10000) < 0.14)
5 tries: 0.1404, 0.1411, 0.1389, 0.1392, 0.1425
try here: https://ideone.com/l/octave (paste code and run)
If it doesn't it's because the simulation was not properly conducted (no code posted, so) or there was no simulation at all, simply a fixed chart to elicit a response (IMO the case here). My point is, there is no need to simulate anything, because there are 2 possible outcomes and the probabilities of each are known beforehand, so it adds no value (it would if you didn't know the odds and had to estimate them, for instance). The whole concept of this thread is just stupid, IMO the OP is just trolling.

People just have a completely understanding of probability.

[JPB]

Are we really discussing between grown people how probability work ?

[exstatic]

Any system can be rigged.

Non asnswer.

[Ed Helicopter Jones]

One time I was bored I wrote a small program to calculate the exact odds, to double check on Tankathon. Their odds check out, here with a little extra precision (per row, red -hot- to blue -cold- for most likely to least likely positions per record):

also did a small program to generate a large number of batches, because Tankathon was not giving me accurate results. Simulating this is pretty fast and cheap, Tankathon adds a lot of effects and delay for dramatic purposes.

Interesting. Finishing 3rd worst means you have only about a 40% chance of a top 3 pick.

[offset formation]

People just have a completely understanding of probability.

Yea, because we're unsure if it takes 10,000 or a million or whatever tries to shake out the stated probabilities at ~14% instead of 13 3% or 15.1% or whatever.

I took statistics in college and got an A but I know plenty of really smart people that struggled with this class.. The NBA draft odds are not something one understands intuitively because you don't get precisely 14% odds. It has to pare down, revert to the mean if you will, over a mul ude of runs. That's all that's being discussed here.

[offset formation]

Technically when you flip a coin, you have a 50% chance for both heads and tails.

That doesn't mean that out of 100 flips, the final count will be 50 and 50.

Well aware of this. Which is why I'm interested as to how many runs it takes to make the mean the most common result.

[scott]

Yea, because we're unsure if it takes 10,000 or a million or whatever tries to shake out the stated probabilities at ~14% instead of 13 3% or 15.1% or whatever.

I took statistics in college and got an A but I know plenty of really smart people that struggled with this class.. The NBA draft odds are not something one understands intuitively because you don't get precisely 14% odds. It has to pare down, revert to the mean if you will, over a mul ude of runs. That's all that's being discussed here.

This is a matter of semantics, but you do get precisely 14% odds, but you only get one chance, meaning the outcome is either 100% or 0%. But the probably is still precisely 14.000000000000000000000000000000%

[scott]

Well aware of this. Which is why I'm interested as to how many runs it takes to make the mean the most common result.

Not sure if what you want to ask is exactly how you have this question worded, but except for possibly a very small number of iterations, the mean will never equal the mode (the most common result) except for those teams with very low odds of moving up (with these odds, only team 13 and 14 made their rounded mean = their mode). The mean is the Expected Position in Ariel's table, whereas the mode (over time) will be the pick in red in Ariel's table.

The mode is not tremendously valuable information, because as has been pointed out, you only get one spin. You can look at Slot 3 and say that the most common result is Pick 6 - but that should not be interpreted as it being most likely we get pick 6 since there is still a 74% chance we do NOT get pick 6. We are far more likely to not get Pick 6 that we are to get Pick 6.