Thread by @jimmysong

1/ Back of the envelope math for doing a 58 block reorg (current confirmations for the tx that took money from binance):

Minimal cost: 58 * 12.5 btc = 725 BTC (assumes every miner would get roughly the same tx fees in the new chain and that 100% of miners go with this scheme)

2/ If 75% of the network going with this scheme, you would need on average 116 blocks to overtake the current chain, or about 1450 BTC worth of mining rewards.
At 60%, this becomes 290 blocks or 3625 BTC. At 55%, 580 blocks/7250 BTC.

3/ 7250 BTC > 7000 BTC, so at minimum, you need more than 55% of all hashing power to agree to reorg the chain. In 58 more blocks, this rises to 60%, in 116 blocks, 65%. In 174 blocks, 70%, 232 blocks/75%. The cost goes up pretty quickly assuming everyone agrees.

4/ There are significant risks also for the miners agreeing to do this as any miner that betrays will significantly raise the risk of wasting hash power that could be productively put to mining the longer chain.

5/ Depending on the premium that a miner would charge for that risk, this would mean a much shorter time than 232 blocks to make that decision. And really, if you do this, you end up taking the money from the thieves and giving it to miners. Binance doesn't benefit that much.

6/ The losses are still at minimum 725 BTC which is still not a small chunk of change. /fin

7/ I lied. Binance is squeezed by time and risk. Every new block is another one that needs to be overtaken and costs 12.5 BTC minimum. Coordinating that quickly with the pools out there is likely not easy. The agreement process alone will take more than a day or 144 blocks. /fin