Disclaimer: This post was originally posted in the Safemoon Investing sub - a place where people can discuss Safemoon-critical information without being censored by mods.

---

Hi everyone. Lately, Coffezilla posted a series of excellent videos on Safemoon, but in the first video of his deep dives he made a very critical error which the Safemoon army pounced on as a way to invalidate all of his claims.

Coffee claimed that in the V1 to V2 migration, the total amount of V1 lost was in the $100m+ range. This number was not even close to correct, I had previously posted my own findings which suggested around $6m, 2 weeks prior to Coffee's video.

Before I get in to it, I want to get everyone up to speed here.

Coffee's number is right and wrong, as explained by a member of Crypto Cynics called Strider. Strider looked at the amount of money that entered the liquidity pools as a whole instead of what was taxed that wasn't meant to be migrated.

While I did challenge these numbers in late March, what none of us really knew at that point is that when anyone migrates to V2, their V1 is still 100% taxed, but they are reimbursed V2 from the deployer wallet. Therefore it does look like over $100m has been taxed. That is true, but most people got their money back by being reimbursed with V2 tokens.

This research wasn't just me, I had a huge amount of help from another member called Bicam, who was just fucking excellent the whole way through. We learned so much, bouncing theories off each other and trying to work it out. The information here is presented by me but it was a joint effort to get there so any credit deserves to be given to Bicam too.

---

---

There's some facts that do need to be laid down first. I'm most interested in this side of Safemoon because I find it really perverse that a company can punitively take their investors and cheerleaders funds, and that the most avid followers of that company will cheer them on. The whole thing is really fucking weird to me, and when Coffee published his video I resolved to find out the correct number, or at least as close as I could get.

---

Method ID's

When a holder has V1, they can do five things with it:

• Migrate it
  
• Transfer it
  
• Sell it
• Buy more
• Swap it
  

Everything but migrating it results in the 100% tax, where the taxed tokens are delivered to the liquidity pool via the Smart Contact. Each of these five methods has a "Method ID" and indeed, selling, swapping and buying have three MethodID's each. As an example:

SwapExactETHforTokens, SwapETHforTokens, SwapETHforExactTokens

The layman way to describe this is when you swap for EXACT, you are specifying exactly what you want to get. It's like a limit order vs a market order. In the interest of thoroughness, all three methods were scanned for transactions and counted for buys, sells and swaps.

---

Time Range

I'm going to be looking from D-Day (December 29, 2021 16:00 UTC) to midnight on April 24th. Tax losses are, unfortunately, ongoing so there is no way to measure the last moment.

---

Pricing

Previously, I simply used the daily close price of Safemoon V1 from CoinMarketCap to determine the price. This was a huge average and wasn't really accurate - the previously quoted $6.3m figure was a result of this, and you'll see how inaccurate it was.

The numbers are therefore the price of the token that was lost at the date and time of transfer.

Bicam figured out a way to query the blockchain to pull the minute-by-minute price of Safemoon. I simply matched the minute of the price to the minute of the transaction in order to determine the USD value of each taxed event. Same again with BNB.

---

Follow Along

We used Dune to query the blockchain as they have a complete record of BSC. If you have any SQL knowledge, you will find this very easy to follow along. Dune is an invaluable resource and I'm already looking at other suspicious moments in Safemoon's history.

When you load Dune, select "Binance Smart Chain" on the left dropdown and then simply copy and paste the provided SQL into the box. Some queries can take up to 30 minutes to run (lots of joins and substring searches)

---

For the Safemooners who might be reading this.

Look, in all honesty, this isn't a nice thing for Safemoon to do. They could've set the tax to 101%, erroring out all transactions on PancakeSwap and Safemoon swap, and then did a snapshot of all holders & balances, then transferred them the equivalent of V2. And that would be fine, you know? Hell, even say everyone with less than 500,000 V1 doesn't get moved over, just to group up the wallets. I would even be ok with that.

But this is your friends, your "sides" money that was token. I've looked at the transactions. It's not all 'arbitrage bots' - for some people, it was their first time ever owning a BSC token and people lost a LOT of money. The following is entirely factual, all caveats are listed, and it's up to you to make up your mind whether you think this is acceptable to have your money taken from you because you didn't follow social media before making a transaction.

---

Transfers - $3,946,944

This is defined as moving Safemoon from A to B, perhaps from TrustWallet to Safemoon wallet, or whatever. This is the bulk of the tax losses.

• 37,475 Taxed Transfers in the Time Range
• 4,159,929,765,592 Safemoon V1 Taxed (yes, 4.16 Trillion)
  
• $3,946,944 in USD value lost.
  

select tx.block_time, tx.hash, tx.from, tx.to, tt.value, (bytea2numeric(substring(tx.data from 37 for 32))-1)/1e9 as fee_transfer from bep20."BEP20_evt_Transfer" tt INNER JOIN bsc.transactions tx on tt.evt_tx_hash = tx.hash where tx."to" = '\x8076c74c5e3f5852037f31ff0093eeb8c8add8d3' --SFM V1 and tx.data::varchar like '%xa9059cbb%' --methodID of a transfer and tx.block_time between '2021-12-29 16:00' and '2022-04-25 00:00' and tx.success = TRUE and (bytea2numeric(substring(tx.data from 37 for 32))-1)/1e9 > 100 order by tx.block_time asc

---

Purchases (SwapETHforTokens) - $842,201

I know it says ETH, but BSC blockchain is a copy of ETH and apparently any reference to swapping a native token for a regular token is called ETH. Roll with it.

• 15,915 taxed transactions
  
• 1,737 BNB lost
  
• $842,201 worth of BNB lost.
  

Remember that when these people selected to purchase Safemoon with BNB, the transaction executed, their BNB was taken from them, and they were given 0.0000001 Safemoon V1. So their BNB was sucked into the Liquidity Pool.

select * from bep20."BEP20_evt_Transfer" tt INNER JOIN bsc.transactions tx on tt.evt_tx_hash = tx.hash where right(tx.data::varchar,8) = 'c8add8d3' --SFM V1 Contract and left(tx.data::varchar, 10) = '\xfb3bdb41' --swapETHForExactTokens and tx.block_time >= '2021-12-29 16:00'

---

Swaps $110,594 (ongoing)

Swaps are a weird one. Many things can be swapped for Safemoon but it was most often things like Tether, BUSD, Dogecoin, wrapped Cardano, Wrapped ETH, Wrapped BTC.

The way to work it out is to look at the first transaction in the chain and then the value at the time. Ignore the value for everything else (a token swap chain can be 6 or more swaps long!)

So it's the value of the input transaction. As the output was zero, the owner lost everything and therefore that's the value lost to tax.

• 1,613 taxed swaps (I only have data till 23rd April)
• $110,594 in tokens lost
  

select tx.block_time, tx.hash, tx.from, tx.to, tt.value, tt.contract_address, cts.name from bep20."BEP20_evt_Transfer" tt INNER JOIN bsc.transactions tx on tt.evt_tx_hash = tx.hash LEFT JOIN bsc.contracts cts on tt.contract_address = cts.address where right(tx.data::varchar,8) = 'c8add8d3' --SFM V1 Contract and left(tx.data::varchar, 10) = '\x8803dbee' --swapTokensForExactTokens and tx.block_time >= '2021-12-29 16:00' and tx.success = TRUE

---

Sales ($0.40)

Yeah. Weird one. Myself and Bicam can't find a dicky bird on these. We've pulled 776 transactions that used a Selling methodID (SwapTokensforETH), but it seems like there were no sales of Safemoon for BNB for anything more than 100,000 tokens.

My theory on this is that something on PancakeSwap / Safemoonswap prevented people from making larger sales due to slippage, but sales of 100,000 or less seem to go through. This is corroborated by the amount of Safemooners who said "You can't get taxed without seeing all these error messages" - maybe they were just looking at the Sells... I don't know.

select tx.block_time, tx.hash, tx.from, tx.to, tx.data, tt.contract_address from bep20."BEP20_evt_Transfer" tt INNER JOIN bsc.transactions tx on tt.evt_tx_hash = tx.hash and contract_address = '\x8076c74c5e3f5852037f31ff0093eeb8c8add8d3' and left(tx.data::varchar, 10) = '\x791ac947' --swapExactTokensForETHSupportingFeeOnTransferTokens and tx.block_time between '2021-12-29 16:00' and '2022-04-25 00:00'

---

---

So that's it.

Total amount taxed from D-day to 25th April:

$4,899,739

I believe this number is the most accurate Tax figure so far, and while I admit there may be more revelations to come, I believe this is the baseline. Every SQL query here gets you a line-by-line transaction log pertaining to that methodID.

Each transaction can therefore be investigated and validated manually.

---

This is the result of about 2 weeks work, and still I don't think I can close the book on it. People are being taxed every day, simply for not keeping up with social media posts from last year. It's unacceptable, and whatever side of the fence you find yourself on, I hope that you will add your voice to the growing roar that demands investor money back.

What Safemoon did was unprecedented - No other Crypto taxed you 100% for simply moving your coins. There was no reasonable expectation that investors would have their funds hijacked by the development team.

For everyone claiming that Safemoon will reimburse these people, I don't know... They seemed quite happy to constantly dip their hands into the Liquidity pool and withdraw tokens. Notice the extreme amounts of liquidity added around December 29th? And... how much has been withdrawn since.