Leadership · Retrospective
What six token-holder votes decided

On 2 September we posted the result of DAO Proposal #6. With 45.63% of the vote, PYM holders chose to raise the cost of crafting PYM from SGEM, our in-game currency, from 2,000 SGEM per 1,000 PYM to 10,000 SGEM per 1,000 PYM. The change took effect on 1 September, the date the proposal had set, once voting closed that day.
It was the sixth decision we had put to token holders since I wrote the first proposal in May 2022. Six is enough to see a pattern.
The six questions
- May 2022. What to do about the 1.5 Million Giveaway after suspected referral bots climbed its leaderboard: cancel it, continue it, or cancel it and pay a smaller amount to the genuine top referrers.
- July 2023. How to split the Anchorland leaderboard rewards in PYM: keep the top 50 as they were, extend rewards down to 300th place, or raise the top 50 and add places 51 to 100.
- September 2023. How long Happy Hour should last. 75.91% chose six hours instead of the existing twelve.
- October to November 2023. Whether Anchorland rewards should be paid in PYM or in SGEM, and how often. 53.48% chose SGEM every 14 days.
- May 2024. How far down the leaderboard rewards should reach. They were extended from the top 50 to the top 100.
- August 2024. The crafting cost above.
What kind of questions they were
Every one of the six was about the settings of the game's economy: how much, how often, to whom and at what cost. Each offered options the team had already judged workable, so whichever won, we could ship it. None asked holders to design a mechanic, approve code or change how a battle resolves.
Those are the questions a vote suits. They are judgements of preference without a single correct answer. A designer can model what a larger reward pool does to the supply of SGEM, but whether players would rather have more winners or bigger prizes is not something a model settles. Token holders also carry much of the cost of the answer, whichever it is.
What the numbers do and do not say
The votes count holders, not players. In the first proposal, each PYM held in a wallet or staked counted as one vote, checked against a snapshot of the chain, and voting cost nothing but a signed message. Most of the later proposals fixed eligibility at a snapshot taken at or before the moment voting opened. Since mid-2023, though, new players have been able to start with an email address and free starter Playermons, without holding any PYM. Their preferences are not in any of these percentages.
A winning share is not always a majority. Proposal #6 was decided by 45.63% across four options. The options were steps on one scale, at 2,000, 5,000, 10,000 and 20,000 SGEM per 1,000 PYM, so the result still says something about where the room sat. Less than half is still less than half, though, and a result announcement should say so in as many words.
Options shape answers. In #6, the description of keeping the current rate was a single line saying it would not address selling pressure. Whoever writes the options sets the agenda, and each option's cost deserves as plain a statement as its benefit. The new rate made PYM scarcer, and it also meant each SGEM now crafts one fifth as much PYM as before. The proposal did tell players to craft any SGEM they held before the change, which is the kind of notice a cost change needs. Even so, a proposal published on 28 August with a snapshot on 30 August mostly counts the people who were already watching.
Incentives tilt too. After #4 we sent participants PYM equal to 1% of their vote as thanks. That encourages turnout, but it pays most to those who already hold most, the same tilt the vote itself has.
What a vote cannot decide
Some decisions that affect our players are not ours at all. On 4 September, Polygon made POL the gas and staking token of Polygon PoS in place of MATIC. Our game runs there, and no PYM vote reaches that far.
Other decisions are ours but do not belong in a voting window of a few days: security fixes, anti-cheat rules and release schedules. Marketplace V3, with batch listing and batch buying, went live on mainnet on 23 September without a vote, and that is how it should be.
Six votes in a little over two years decided mostly where the economy's dials sit. The vote itself is the last step. The leadership work comes before it: choosing a question that suits a vote, writing options that state their costs, giving people time to take part, and publishing the result with its number, including when that number is under half.