I’m not entirely sure I understand your question, so pardon me if this doesn’t answer what you’re asking.
When using a “basic dumb trainer” and a speed or speed/cadence sensor, you are telling Zwift to derive your power output with zPower. What this means is that Zwift uses a predefined power curve for the trainer you selected in the setup. That power curve tells the software “this trainer requires X watts to go Y speed.” In other words, the trainer manufacturer or someone else has tested the trainer and provided a table of values that relate a given wheel speed to a certain power requirement. Zwift then uses that table in conjunction with your speed sensor to calculate your power output. The software sees a wheel speed of, for example, 20 miles an hour, looks at the power curve table, and sees that for this particular trainer that speed corresponds to an output power, for example maybe 200 watts. That derived power number is then plugged into the game’s physics engine and, along with the rider weight you entered, the type of bike and wheels you selected, the climb gradient of whatever section of course you’re on, etc is used to calculate your in-game speed. Because of the physics engine, actual wheel speed from your sensor and game speed are almost never the same thing.
Think about it this way: if you were putting out a steady 200 watts outdoors on a flat road with no wind, what would your speed be? If you were putting out that same steady 200 watts on that same flat road, only this time with a 15 mph headwind, would your speed be the same? How about that same 200 watts on a steady 6% climb? In the real world, these kinds of variables affect your speed on the road, so they affect your wheel speed. In Zwift, these variables (wind speed, grade, etc) only exist in the game, so your speed is only affected in the game.
Hope this helps.