There are a few more related topics to discuss, as well. For example, lower and mid-rpm throttle response is likely with EFI. Stated another way, such equipped engines tend to accelerate through their rpm ranges more quickly and waste less fuel in the process. Based on a more uniform and consistent burn (cycle-to-cycle and cylinder-to-cylinder), the development of combustion pressure in an EFI engine is also characteristic of how engines perform when provided air/fuel charges that are well mixed. Also keep in mind that as atomization efficiency improves, well-blended air/fuel charges can negotiate changes in induction path direction (and inlet pressure variations associated with unsteady, non-homogeneous flow) with minimal air/fuel separation.
Perhaps a more obscure issue relates to how well an engine's torque output (curve) can approximate its volumetric efficiency curve when using EFI. In theory, minus any pumping losses, torque curves should closely match v.e. curves, primarily because if an engine is doing a good job of ingesting air, the potential for achieving equally high levels of output is present. However, if the fuel being delivered during a given operational range of rpm is poorly atomized, air/fuel charges will reflect this inefficiency, resulting in less than optimum power. Such inefficiency also encourages problems associated with excessive air and fuel separation, largely mechanical, along inlet paths and in the combustion space. EFI can have a direct and favorable impact on these conditions, as witnessed (again) by the example data provided.
Of course, there's always the on-track fuel economy issue, becoming increasingly important as dictated by fuel costs. Bottom-line, if it's possible to make more power with the same amount of fuel delivered to the combustion space (compared to carburetion), less is wasted in conjunction with increased on-track power; clearly a healthy combination. EFI is one approach to accomplishing that objective.
Now all this isn't to say carburetors are unacceptable. For many reasons that include rules requirements, convenience, cost, and prior experience at the racer level, carburetors are woven into the fabric of circle-track racing and will likely remain so for a long time. But if you look back over their history in motorsports, much time and expense have been devoted to improving their ability to supply air/fuel charges with features that include good atomization efficiency.
Aside from companion problems related to how wet flow is handled downstream of any "mixing" device or system (carburetors or EFI), the probability of raising net atomization efficiency is greater if the wet flow begins with well-mixed air/fuel charges based on improvements in mixture quality as a function of enhanced atomization. This month, we've had an opportunity to link theory with supporting data. That's a nice and meaningful package.