Restrain yourselves!!

So, what's been happening in Free Agency today? Let's have a look: -- Erm, nothing, zero, zilch. Oh dear. In that case, now is as good a time as any to investigate a mystery that's probably been bugging some of you for the last few days.... while others, probably don't care. What does Belichicks decision to go for it on 4th and 2 against the Colts have to do with Temporary Restraining Orders? Well, I'll tell you. If you don't own a copy, go out and buy Gavin De Beckers book "The Gift of Fear". Buy a copy for your wife or girlfriend as well. Buy a copy for any brothers and sisters you have. Buy a copy for any sons and daughters you have (I mean it). If you have enough spare change, buy this book for everybody you can and tell the rest that it's a must read. I don't want to dwell too much on the book, but to give you a brief summary, De Becker is an expert in security and details in this book many of the main topics he's encountered through his career, sharing his insight as to how his company deals with various situations and persons, from stalkers to violent children to lovers that won't let go, and even mentions assassins. Becker demonstrates the difference between fear and worry and covers so many other interesting topics. A must read. And thus we come to the bizarre correlation between the Head Coach of the New England Patriots, Game Theory, and a court order used to ward off the unwanted attention of a fan/girlfriend/boyfriend/husband/wife etc. So to start what is rapidly becoming a muddle in my mind, basically a temporary restraining order is a court order used to restrict the access of one individual to another, usually setting limits on distance, time, places etc. And as we find out in De Beckers book, they generally do one of two things: 1) Serve the purpose for which they were intended, admirably. 2) Make a bad situation much worse. The key factor is to look at the history of the person being placed under the order. Normally harmless, law abiding individuals will tend to get the message and cease their previously relentless pursuit. Those with a past history of violence and threats to use violence will likely be pushed over the edge by the order, now feeling that the target of their unwanted attention and pursuit has won a victory over them and that they have no other options left on the table. Still with me? Ok, so why is this important? Because it affects the ways we look at statistics. We can call up some arbitrary number (being that I lack the actual numbers in front of me) and say that every year there are (X) amount of restraining orders issued and that in (Y) number of cases, the restraining order makes the situation worse and ends with the death of the victim of the pursuit. From this, you could quite easily say that, therefore, if you take out a restraining order against someone, the percentage chance that they will end up trying to kill you is (Z) or something like 1 in ABC chance of it happening. But the reality is that we know with quite a high degree of accuracy and consistency who will become violent and who wont. We can look at past form and analyse any number of variables to give a person a good idea of whether the case will be successful or not. Generalised percentages are of no use here. And now we come to the Belichick part. After his now much maligned decision we were informed repeatedly that the percentages were in his favour. I heard everything ranging from 65% (the most consistent number) all the way up to 80% (god knows how someone came to that conclusion). People went statistics crazy, trying to transpose all kinds of numbers onto the situation. The only one percentage I didn't actually hear was 50%, which was the actual Patriots record on fourth downs that season prior to that attempt. All of this is a prime example of trying to impose arbitrary numbers and percentages onto a situation that has many variables to it. In the same way that we can look at whether a restraining order is a good idea and assess all the variables involved without having to lump everybodies case into some fictitious percentage, we can also look at thing like fourth downs and make decisions without having to rely on calculators. We can look at the Patriots previous track record on fourth downs that season (many people were looking at the Patriots under Belichick, The Patriots with Brady etc, anything except the Patriots that season with that personnel), we can look at the Colts ability to stop on 4th down, we can look at how well the Patriots special teams had performed against the Colts special teams and we can look at the Patriots defense versus the Colts offense (many people glossed over the fact that the Colts last drive ended with a score only after a huge chunk of yards were gained on a highly dubious pass interference call and that on the drive prior to that, King Manning threw a pick). I think too readily now, coaches seem to be turning to percentages and people with degrees in applied Mathematics to make decisions for them, instead of trusting their gut instincts and looking at the facts that actually present themselves. Would the Maths people who were so confident in Brady still be as confident if they were facing the Steelers, who had a near 80% success rate at stopping offenses on 4th down (to that point in the season)? Or would we still be lead to believe that Brady was 65% all the way? You cant safely dish out a Temporary Restraining Order without first looking at and understanding the facts of the individual case and neither, I propose, can you go for it on 4th down without first looking at and understanding the facts of the game in hand. Hopefully that's all clear now. If not, feel free to let me know in the comments section below. But for now, thanks for your time and at some point over the weekend I'd like to convince you that those high ticket prices you pay to see games and the expensive beer you're forced to buy at the game is all for your own good. Have a nice day.