When Statistics Are Not Helpful by Dave Cameron July 25, 2011 Here on FanGraphs, we talk a lot about probability and likely outcomes. When making a decision, we think it’s generally wise to understand historical precedent, and to learn from history rather than repeat it. But, there are times in life that you’re not making a decision, and knowledge of the probability of outcomes just doesn’t help at all. You are just rooting for one specific result, even if you don’t have any control over whether it occurs or not. I’m now in one of those situations. Last week, I was informed that I have Acute Myeloid Leukemia, a particularly nasty member of the cancer family. History has given my doctors all kinds of data about cure rates and life expectancy, and statistical analysis is helping them decide just what kind of chemotherapy I’ll be taking in a few hours, which I’m really thankful for. But really, those numbers do nothing for me. I’m not going to be making very many decisions over the next few months. I’m just going to be rooting like crazy for the drugs to work. I need reasons for hope, and I won’t find much of that in the harshness of raw data. Data isn’t always what is needed. If you’re a Pirates fan right now, does it help you at all to know that your team probably won’t keep this up? You’re not going to be making any decisions that will change the outcome anyway, so why not root for the outcome you want, even if it isn’t statistically probable? Thats what I’m going to spend the next few months (and years, in reality) doing. Save the odds for the doctors; I’m planning on living a long time. I’m planning on beating this thing. I’m planning on watching the Mariners win a game, and at this rate, that might take years. I want to be around to see it, though, and I just don’t care what the odds say is likely. For the rest of 2011, I’m unsure of what my involvement here will be. The hospital has wi-fi, so if I feel good, I might write ten posts a day. If the chemo sucks, you might not hear from me for a few weeks. At this point, I just don’t know what is going to happen, but I know the outcome I want, and the fact that the data suggests it may not happen is irrelevant to me. Statistics can be powerful, useful tools, and at times, they can be critical to understanding what to do. Other times, though, they’re useless, and so, for this situation, I say screw the data; I choose hope instead. I know many of you are going to want to know how you can help. For now, I’ll just ask that you strongly consider donating both blood and platelets to the Red Cross – they have a critically low supply of both at the moment. Thankfully, my wife is an oncology PA; we have great health insurance and are in the trusted care of her friends and coworkers, so financial assistance isn’t needed at this time. If that changes, I’ll let everyone know, but for now, send prayers in lieu of cash. See you all when I can. Don’t get too used to not having me around.