The Framework: Over Long Horizons - Free Wins In 2019, Schwab cut commissions/fees to $0 and td ameritrade/etrade matched within days; the fee model collapsed and flow migrated to the lowest all-in cost. What happened is (1) they compressed the entire industry's fee model and (2) over time, they captured the majority of flow. Why? Because over a long enough period of time people prefer free trading over fee tiers. This is the direction I believe we will move with regards to perp DEXs. So here we are today: a plethora of perp DEXs that are largely the same, with Hyperliquid at the top and many hoping to just be considered. If you assume traders prefer free over fee then you need to look for DEXs that can go zero fee while having a profitable business. How do you figure out which ones can actually run zero fees? The ones that can offer benign flow to MMs/HFTs. How do teams offer benign flow? Flow segmentation via RPI (retail price improvement) is a great option. I won’t go into detail on RPI here, I covered it at length in my Delphi report (comment if you want access). Traders pay all-in cost (fees + spread + slippage + rejects). Without curated retail flow, makers must widen or over-hedge to survive, so any fee cut you hand to takers comes back as slippage and worse fills. When makers know which lane they’re filling (and that it isn’t latency-arb), they post tighter spreads and firm size, cancels drop, realized slippage improves, and now zero-fee can stand on its own because the maker pnl isn’t the subsidy. This is provable when looking at DEXs spreads and effective trade cost. Flow segmentation allows for free trading and tighter spreads. None of this is to FUD HL. HL is clearly the top DEX today. The point is that if you believe “free wins” in the long run, the current design caps upside: a cancel-priority, lit-only, single-lane book tends to create a last-look-like dynamic, taxing takers via slippage and discouraging size over time. It also makes durable zero-fee nearly impossible because you can’t scale unless you can guarantee the quality of the flow you’re selling to MMs. The only way you can guarantee flow in their model is if you slow particular orders down (ie speed bumps). It is possible but it now means they need to have this logic in the block-building, which is unnecessary at that level. So with this framework that free venues will win users over time, there are currently only two DEXs leaning into this (@tradeparadex & @Lighter_xyz) and each choosing different ways to provide this free trading experience while generating sustainable revenues. I expect fees to be compressed across the board over time and I hope that HL considers new ways to curate flow to run this PFOF business model well without sacrificing meaningful revenues.