You don't get it and I don't care enough to explain that image. You said spreads with different strike performed differently... well, no shit.
Yes that's the tricky part of options trading. Everything is priced according to implied volatility but your profitability is determined by real volatility and how different it is from implied volatility.
So these are the variations of a box spread? Anyway the topic here is about how a credit spread will outperform a debit spread for the same directional bias.
bc the vol surface isn’t flat. Same strike ps and cs are governed by the box arb. An atm cs is the revenue side of the fly on index, provided you are accounting for the forward (atm basis for both). Why? bc upside strikes are -skew on index.
Take an equidistant otm cs and ps on index, but within say 30-delta. Equidistant priced against the synthetic at the expiration in question. On index the cs will trade at a larger figure than the ps. The box arb reflects the synthetic of EITHER spread, not the relationship between an otm cs and ps. in singles; puts are the revenue side. in verts; cs are the revenue side. it is NOT governed by any arbitrage constraints (ostensibly would be arbed away). It’s the origin of the term “risk” reversal. an atm synthetic long or short is the forward. Long c, short put at x. Long the synthetic and short spot = reversal. Separate strikes and you add risk, and skew, thus a “risk” reversal. Skew cannot be arbed away hence the skew/convex vol-surface (smile).
I'm accounting for either direction where the short call vertical (index) will result in better gains and less losses.