What Are the Odds?

No, Top Shot isn't rigged

As you may have heard, there was a fair bit of community uproar yesterday afternoon after Andrew Brown, collector iracer77 on Top Shot, pulled the #1 Chris Paul Rare Conference Finals Moment. Pulling this Moment from a pack didn’t throw the Top Shot Twitterverse upside down on its own - it was the fact that this less than 24 hours after Andrew was selected to attend Game 5 of the NBA Finals in the Top Shot Suite, all expenses paid.

“Rigged!”

“That’s the final straw - I’m leaving Top Shot! What a scam”

Welcome to the beautiful world of my Twitter DM’s. Now, if you’re new around here, let me get you up to speed. It seems as though I’ve carved out a rather odd part of the Top Shot community - the data guy that digs into mysteries and uncovers the truth. Whether that’s uncovering an elaborate multi-accounting scheme, or deciphering how a Challenge Reward went awry, I’ve always enjoyed putting the puzzle pieces together to tell an interesting story.

Based on everything I know about Top Shot, I do not believe for a second that this was intentional or rigged. Why is that? Come right this way.

Let’s Talk Probabilities

Without a doubt, the sequence of events that unfolded yesterday had an incredibly low probability of occurring. How low? We can do some simple math to answer that question.

To start things off, what was the probability of iracer77 being selected as one of the Game 5 Top Shot Suite winners? Just under 3,000 Collectors had a total of 6,061 entries into the giveaway. iracer77 had 5 entries of their own, and with 8 lucky Collectors being selected, he had a 0.66% chance of packing their bags to Phoenix.

Next up, what were the odds of landing a Rare And Then There Were Four pack? Qualifying for all three of the tiered drops, iracer77 had three shots at landing a pack at the following probabilities:

• Priority Queue 1: 10k in queue, 2k packs —> 20%

• Priority Queue 2: 17k in queue, 2k packs —> 11.8%

• General Queue: 89k in queue, 2k packs —> 2.2%

Combine these three queues, and iracer77 had a 31% chance of ripping open a Rare pack.

Lastly, I’m going to take the creative liberty of determining the odds of iracer77 landing a #1 or jersey match for any of the Rare moments in the pack. With a mint size of 750 Moments, and 2 “winners” here, he had a 0.27% chance of landing a serial number that would break Twitter.

So what’s the probability of all three of these independent events occurring? We can multiply the three probabilities together and find that iracer77 had a 0.0006% chance of having all three of these stars align - in other words, a 1 in ~166k shot.

There’s no arguing that this was statistically unlikely, but unlikely doesn’t equate to impossible. To put these odds into perspective, I want to have you think about your odds of being dealt this exact poker hand shown below twice in a row.

You’re looking at a 1 in 1.75 Million chance, odds that are over 10x less likely than what happened to Andrew. How about winning the Mega Millions lottery? You’re looking at a 1 in 30 Million chance, 180x less likely.

Top Shot’s Motivations

Putting the data aside for a second, I want us to think through Top Shot’s motivations. If we put on our tin foil hats and pretend that this was rigged, what’s in it for Top Shot? Yes, Andrew is a volunteer Discord moderator, but do we really think Dapper Labs would go through such great lengths to reward a volunteer? And if we pretend there was intent, don’t we think the team would be smarter than to stuff Andrew’s pack with the #1 Serial number? Surely a #5 would’ve gotten the job done. I’m not buying it.

I’ve also heard that it could be collusion with a Top Shot engineer that could just “tweak the algorithm” to make all of this happen. As someone who works as a Data Scientist writing production-level code, let me tell you that it’s not as simple as sneaking in a line of code when nobody is looking. Software teams have thorough processes for every new line of code that’s merged into the product’s codebase that includes peer review from fellow teammates. Multiple engineers were in on this, and felt it was worth the risk of likely losing their job? Again, I’m not buying it.

And for those of you in the back pointing out that Andrew won a recent Showcase Quest, and that these winners might be hand-picked by the Top Shot team? I took the time to read through past Showcase Quest winners, and considering Top Shot user Your_Mom_69 (wish I could say I was making that up) won the latest Showcase Quest, I am - yet again - not buying it.

Final Thoughts

I understand that some folks can be upset when someone else has a statistically lucky experience, but rushing to grab the pitchforks at every twist and turn is an unhealthy environment. Shouldn’t we be celebrating that we’re part of a community where one lucky collector had an incredibly lucky two-day run? It’s unfortunate to see the community create a situation where a collector like Andrew is now hesitant to participate in a Showcase Quest in fear that he’ll win - how backwards is that?

Lastly, while I wish I had even more concrete evidence to prove that this was not an elaborate plan to reward a volunteer Discord moderator, there are still portions of the internal Top Shot mechanisms that remain in a black box. We can look at the Flow Blockchain and see what comes out of this black box, but we don’t know what’s going on inside of it.

For now, some aspects of Top Shot such as queue randomization and Challenge Reward distribution aren’t visible in plain sight. While other NFT projects enable Collectors to see similar mechanisms in their smart contracts, it’s not the case with Top Shot. Do I bring this up because I’m skeptical? Absolutely not. I’m simply a Top Shot community member who would want to have even more ways to prove why this sequence of events wasn’t rigged. For now you’ll just have to take my word for it.

That’s all for this week! As always, I hope you enjoyed this article. Want to support the newsletter? You can help with the late nights and early mornings by buying me a coffee ☕️