Great question and exceellent reply. Ive been doing a lot of work in this area trying to find true Gex and can confirm all the services you mentioned, none of them use true Gex. Spotgamma is always held as the prime example but they used a form of implied direction which again is not exact. Excellent little video from Dan at Volsignals where he explains in simple terms what I found. https://www.youtube.com/watch?v=8tk6w0K9AIU
Hello! You are using naive GEX assumptions correct? Don’t you think that is far from the actual reality and can mislead specially a 0dte level analysis?
Spotgamma does not have real costumer positioning from CBOE, like other platforms. So they also make assumptions on the side of the option.
Great question — that is the biggest methodological weakness in the post and I should have been sharper about it.
Yes, the GEX framework as presented uses naive assumptions: dealers are assumed short whatever customers are long, inferred from open interest alone. No one outside the clearing firms (OCC) and some CBOE data licensees actually knows who's on which side of the trade. SpotGamma, Squeezemetrics, GammaLab — they're all reverse-engineering dealer positioning from public OI and volume data with assumptions about customer vs. market maker flow (at least that's my understanding).
So this creates two problems IMO:
1. Side-of-trade ambiguity. If a hedge fund sells puts as a yield strategy (not buying for protection), the dealer is long those puts — the sign flips. For single names this is especially problematic since institutional hedging, structured product issuance, and vol overlay programs all create positions where the naive assumption is wrong. For SPX/SPY the assumption is more defensible because the dominant flow is genuinely customer buying of puts for protection and calls for upside — but it's still an approximation.
2. 0DTE breaks the model. Classic GEX uses end-of-day OI. 0DTE options open and expire within the session — they never show up in OI snapshots but generate enormous intraday hedging flow. You could have a completely different intraday GEX profile than the EOD snapshot suggests. Some estimates put 0DTE at 50%+ of SPX options volume, so this isn't a rounding error.
So what I'd argue still holds: the direction of the relationship (positive aggregate GEX --> lower realized vol, negative --> higher) is robust even with noisy GEX estimates, because it's driven by the mechanical delta-hedging feedback loop, not by the precision of the positioning estimate, if that makes sense.
So you don't need to know the exact GEX level — you basically just need to know the regime (positive vs. negative), and in my experience even crude proxies get that right most of the time.
But for 0DTE-level analysis, or for anything claiming precision at the strike level? You're totally right — naive GEX is insufficient. In fact I'd argue that the platforms that claim strike-level precision without actual CBOE customer positioning data are selling more confidence than the methodology probably supports IMHO.
I'll flag this more explicitly in the post. Thanks for flagging!!
Have followied several Substack authors who write daily about ODTE options, GEX etc. (as it relates to US stock indices) and have been trying to get a better understanding of how hedging by market makers impacts and influences the spot index price, all with a view to trying to apply the same analysis to GEX, Gamma flip etc. for individual stocks. Would you please be able to point me in the direction of articles/information that would help me learn more about this? Thank you very much!
Great question and exceellent reply. Ive been doing a lot of work in this area trying to find true Gex and can confirm all the services you mentioned, none of them use true Gex. Spotgamma is always held as the prime example but they used a form of implied direction which again is not exact. Excellent little video from Dan at Volsignals where he explains in simple terms what I found. https://www.youtube.com/watch?v=8tk6w0K9AIU
Hello! You are using naive GEX assumptions correct? Don’t you think that is far from the actual reality and can mislead specially a 0dte level analysis?
Spotgamma does not have real costumer positioning from CBOE, like other platforms. So they also make assumptions on the side of the option.
Great question — that is the biggest methodological weakness in the post and I should have been sharper about it.
Yes, the GEX framework as presented uses naive assumptions: dealers are assumed short whatever customers are long, inferred from open interest alone. No one outside the clearing firms (OCC) and some CBOE data licensees actually knows who's on which side of the trade. SpotGamma, Squeezemetrics, GammaLab — they're all reverse-engineering dealer positioning from public OI and volume data with assumptions about customer vs. market maker flow (at least that's my understanding).
So this creates two problems IMO:
1. Side-of-trade ambiguity. If a hedge fund sells puts as a yield strategy (not buying for protection), the dealer is long those puts — the sign flips. For single names this is especially problematic since institutional hedging, structured product issuance, and vol overlay programs all create positions where the naive assumption is wrong. For SPX/SPY the assumption is more defensible because the dominant flow is genuinely customer buying of puts for protection and calls for upside — but it's still an approximation.
2. 0DTE breaks the model. Classic GEX uses end-of-day OI. 0DTE options open and expire within the session — they never show up in OI snapshots but generate enormous intraday hedging flow. You could have a completely different intraday GEX profile than the EOD snapshot suggests. Some estimates put 0DTE at 50%+ of SPX options volume, so this isn't a rounding error.
So what I'd argue still holds: the direction of the relationship (positive aggregate GEX --> lower realized vol, negative --> higher) is robust even with noisy GEX estimates, because it's driven by the mechanical delta-hedging feedback loop, not by the precision of the positioning estimate, if that makes sense.
So you don't need to know the exact GEX level — you basically just need to know the regime (positive vs. negative), and in my experience even crude proxies get that right most of the time.
But for 0DTE-level analysis, or for anything claiming precision at the strike level? You're totally right — naive GEX is insufficient. In fact I'd argue that the platforms that claim strike-level precision without actual CBOE customer positioning data are selling more confidence than the methodology probably supports IMHO.
I'll flag this more explicitly in the post. Thanks for flagging!!
Have followied several Substack authors who write daily about ODTE options, GEX etc. (as it relates to US stock indices) and have been trying to get a better understanding of how hedging by market makers impacts and influences the spot index price, all with a view to trying to apply the same analysis to GEX, Gamma flip etc. for individual stocks. Would you please be able to point me in the direction of articles/information that would help me learn more about this? Thank you very much!