Within the lab, exercise scientists and trained professionals will set up athletes on a piece of stationary exercise equipment, such as a bike or treadmill. Using a specially-designed assessment tool, athletes place a heart rate monitor around their chest and a mask fully around their mouth and nose. These devices connect to a machine to collect data - both heart rate, oxygen intake, and carbon dioxide exhalation. After a period of calibration, the athlete will begin exercise at a low level of intensity, gradually increasing intensity at regular intervals (usually every 1-3 minutes). The measured VO2 rises accordingly until the athlete reaches their max, upon which the machine will indicate a plateau.
While this is all well and good, I've yet to see Dave Castro put "masked breathing run hooked up to a machine for distance" in the Open. So can it really serve as a test for CrossFit?
Other, more practical methods do exist. For example, you could estimate VO2 max using a treadmill or bike. The Bruce Treadmill method involves seven stages of increased running speed and incline levels every three minutes. The test continues until exhaustion, and an estimate of VO2 max is determined accordingly. That protocol can be found here.
You can also use a Wattbike to do the same for cycling, which could be important for those of us who frequent the Assault Bike. That protocol can be found here.
Regardless, CrossFit is so much more than running or biking. By definition, it requires fitness across multiple codes - anaerobic endurance, aerobic endurance, strength, power, speed, flexibility, and skill. While scoring well on the above VO2 max assessments indicates strong aerobic endurance, it tells you nothing about your ability to snatch 225lb or climb a rope. However, there is one other hidden measure within these tests that does correlate with success in CrossFit - the lactate threshold.