Notes on how Ponbond works

pIDX is a synthetic index. Holding one does not make you the owner of anything sitting in a vault. It is a claim whose price is meant to track a fixed basket, and the only thing standing behind that claim is collateral somebody locked up in order to mint it. If you are going to hold a synthetic, the collateral rule is the part worth reading, not the basket.

Overcollateralised means the lock is worth more than the claim it issued. pIDX is designed to open at 300 percent, so three dollars of collateral for every dollar of exposure minted, and to become liquidatable at 225 percent. The gap between those two numbers is the whole design. It is the room the basket has to move against the minter before anybody needs to act, and it is why a liquidation can usually be settled while there is still more collateral than debt. A system that let people mint at 100 percent would have no room at all, and the first bad hour would leave claims with nothing behind them.

Both figures are parameters, not results. The mint, burn and liquidation paths are not deployed, nothing has been minted against them, and no ratio is being enforced today, which is why the Indexes page shows a checklist of what has to happen before minting opens rather than a market you can trade. What is written here is the rule the contract is built around, and it is worth understanding before it goes live rather than after.

What overcollateralised does not mean is safe. Collateral can fall on the same day the basket rises, which is the worst case for a minter and the case where the buffer disappears fastest. A stalled chain, a lagging price feed, or a market that gaps through the level can each leave a position underwater before the ratio can be enforced. And the enforcement is code: Ponbond's contracts are unaudited, which the terms state plainly rather than burying in a footnote. Read the ratio as a buffer that was designed on purpose, not as a guarantee that somebody is standing behind it.

A Ponbond profile is nothing more exotic than a list of target percentages across the five reserve assets, and those targets always add to 100. A USDG deposit into a profile is split along those targets on the way in. That is the easy half, and it is the half every product gets right.

The hard half is what happens next. The five assets move at different speeds, so a month later the mix is no longer the mix you chose. Whatever ran hottest is now a larger share of your position than you asked for, which means your exposure to it is largest exactly when it is most expensive. Rebalancing is the correction. Trim what grew past its target, top up what fell below it, back to the numbers printed on the card. It is mechanical. It predicts nothing. It only refuses to let the mix drift into a concentration nobody chose.

This is also the part people are worst at doing by hand, because doing it correctly always feels wrong. You sell a slice of the thing that is working to buy more of the thing that is not. Handing that to code removes the argument, but it does not make it free. Every rebalance is a trade, trades cost something, and the management fee is 0.35 percent a year on top of that. In a stretch where one asset beats everything else for years, a rebalanced mix finishes behind a concentrated bet on that asset, and it will feel like a mistake the entire time. That is the trade being made. It should be a decision you took, not a surprise you discover later. Three of the four profiles hold their weights exactly as printed; the adaptive one is the exception and moves its own targets over time, which the app labels where it applies.

A headline APY is a rate quoted at a moment. It reports what something paid over a recent window, annualised, and it says almost nothing about what your principal is standing on. Two products can advertise the same figure while one earns it from short government paper and the other from tokens minted specifically to make the figure exist. The number is identical. The thing you are holding is not.

A portfolio starts from the other end of the question. Not what does this pay, but what do I own, and what happens to the rest of it when one piece falls. The Ponbond reserve is five holdings with five different jobs: a broad stock index, a tech index, short dated treasuries, silver, and cash. They do not all fall on the same day for the same reason, and that lack of agreement is the entire point of holding more than one of them. A portfolio does not remove risk. It stops a single bad outcome from being the only outcome that matters.

So there is no APY on the front of this app, because we would have to invent one. What the app shows instead is what a given mix would have returned over a window of one, three or five past calendar years, using each asset's published annual returns, with the fact sheet behind every figure linked so you can check the inputs yourself. Cash is the one line with nothing to link, because it is held at par at zero percent and the table says so. That is arithmetic applied to history. It is not a projection, past years do not repeat on request, and even the longest window we have data for is short enough that one unusual year can dominate the whole result. That is why the worst year in the window is displayed next to the result rather than tucked away underneath it.