Tougher Extra Points: What’s the Difference?

The NFL opted this preseason to move extra point attempts back to the 15 yardline, 13 yards back from the traditional 2 yardline. It’s no secret that extra points after touchdowns have been essentially automatic over the years. In the 2013 regular season, a grand total of 5 extra points were missed on 1267 extra point attempts. That’s a 99.6% conversion rate. This preseason has given us a total of 196 attempted extra points thus far, which is by no means a large data sample compared to a full season, but it’s something upon which we can at least make some premature projections.

8 extra points have been missed already this preseason – 3 more than were missed during the entirety of the 2013 season, for those who struggle with subtraction. That’s a 95.9% conversion percentage, which may not seem significant, but such a difference played out over an entire season can make a major difference. Only 1215 of the 1267 field goals (52 misses, instead of only 5) would have been made with such a conversion rate in 2013.

Before we even proceed, let me again mention that this data is not necessarily reliable. It’s a smaller sample size, and the preseason sees both second team action and kicker position battles during first team repetitions. What this means is that the team’s best kicking talent is not always taking the extra point attempts during the preseason. That being said, the opposition’s special teams personnel won’t always be up to par either, meaning less effective pressure will be on the kicker on many attempts. There’s plenty of reason, then, to believe the conversion rate will be different in the regular season.

To illustrate this, let’s look at the full 2013 season stats for field goal attempts between 30-39 yards, since extra point attempts under the new rules fall within that range – 33 yards. 295 such field goals were attempted, with 30 misses. This is actually a lower percentage, at 89.8%. Obviously a 39 yard field goal is more difficult than a 33 yard field goal. So we can’t expect the conversion rate to be 95.9% or 89.8%, but we can reasonably expect it to fall in between those two numbers.

Instead of trying to find some arbitrary weighted average, I’ll look at the two extremes and the scenarios for both. 89.8% of the 1267 extra point attempts would be 1138 extra points made, a relatively shocking 129 misses. That’s about 4 misses per team if they were all equally allotted (which they wouldn’t be). What, though, does this mean? In a season featuring 512 games, 52 misses would mean each game would have approximately at 10.2% chance of having a missed extra point (obtained simply by dividing the two figures – obviously a very rough estimate) assuming each game is weighted equally (an absurd assumption, but it would be quite the undertaking for me to weight each game in such a way by kicker’s individual conversion rates, and I thought up the concept for this blog post probably half an hour ago). At 129 misses, this figure inflates to 25.2%. These are both significantly high chances compared to what we’re used to.

When, though, will a missed extra point actually matter? Only in games with very small margins of victory. It’s a very direct impact if the margin is one point, obviously, a missed or made extra point is quite literally the difference of the game. There were 13 such games in 2013, which is a paltry percentage of all games played. Basic probability operations can lead us to a rough percentage chance that one of these games would feature a missed extra point under the new rules (under either scenario). The chance that NONE of the games would feature a missed extra point is (1-.102)^13 or (1-.252)^13. These expressions evaluate to 24.7% and 2.3%, respectively. Therefore, the very broad percentage range that at least one of these games would have featured a missed extra point under this preseason’s extra point conditions is 75.3-97.7%.

Keep in mind that this is all theoretical speculation. Now that it’s in mind, let’s make the speculations even more specific and ridiculous. Some of the games that were decided by one point: Chiefs-Cowboys week 2, won by the Chiefs, 17-16. If KC misses that extra point and the Cowboys win in OT, the ‘Boys are just one game behind Philly for a playoff spot at season’s end. Who knows how that may have changed things down the stretch? That also kills the early-season perception of the Chiefs as an elite AFC team, which everybody saw through by the end of the season. Even more entertaining: Week 14, Patriots 27, Browns 26. If the infallible Stephen Gostkowski misses an extra point in this game and the Browns win in OT, there’s a 3-way tie at 11-5 for the AFC’s second seed between the Patriots, Bengals, and Colts. The Bengals, in that case, would have earned the AFC’s second seed and a first-round playoff bye on the merit of beating both the Patriots and the Colts during the season.

The what-if game is certainly fun, but obviously these are far-reaching hypothetical scenarios. What’s perhaps more applicable is more common margins of victory – specifically, margins of around one touchdown or less. A singular missed extra point, or even a previously-unfathomable-but-now-simply-surprising two missed extra points, will not directly impact the game’s result. That being said, differences of one or two points may affect late-game coaching strategies. Being down by 4 points instead of 3 before a team’s final possession leads to significantly different strategies. Having to score a touchdown is obviously much more difficult than getting within 30 yards of the end zone and kicking a field goal. Being down by 8 or 9 points instead of 7 in the same situation is even more hopeless – a touchdown with a two point conversion is a very low-percentage situation, and there is no 9-point play in the NFL.

Extra point misses will be up significantly. These misses will have implications, and it will be very entertaining to watch these implications play out. I’ll revisit this post by the end of the season and see if the projections are at all accurate.



And Now, Something Different

While I’m passionately interested in Bitcoin, one of my other primary interests is data science. While I’m not an economist by trade, I often find myself skeptical when a blog or other publication throws numbers at us and expects us to read the numbers the way they did. Such an article popped up on BostInno relatively recently, discussing a recent interactive map published by the Bureau of Labor Statistics. The map visualized average weekly salary in 2013 by county, and it showed about what you would expect. Here’s a shot of the map:

Thanks to the Bureau of Labor Statistics!

Thanks to the Bureau of Labor Statistics!

Darker means higher salary, and the map divides the country into what appears to be quintiles. The usual suspects are the darkest-colored, from Massachusetts to Maryland to Alaska. BostInno already mentions that cost of living has to be part of the equation, but doesn’t particularly go into detail on this point. Everyone knows rent is sky-high in more wealthy areas, such as New York and Silicon Valley, but I thought it would be interesting to see how this map would look if it were somehow adjusted for cost of living. Is the “adjusted salary”, so to speak, of wealthy counties as high?

What this boils down to is whether higher salary actually translates to higher living. Obviously higher salary and higher cost of living are anything but mutually exclusive, but it’s not as if there’s a linear relationship between the two. Thanks to a generous discount by the folks over at The Council for Community and Economic Research, I was able to access cost of living data at the county level for 2013. They provided a Cost of Living Index for each county, with the average cost of living having an index of 100. Greater costs of living lead to a greater index. Indices ranged from 84.3 (Zapata County, TX) to 188.3 (Kings County, NY). This is an interesting skew, suggesting that costs deviated right from the average far more than they deviated left.

A simple operation to see if and where in the US higher salary means higher living is to divide the county salary by its cost of living index. This gave me a generally single or double digit number that I’ll call “adjusted salary” for our purposes. The majority of the counties fell between 6 and 8 for adjusted salary. Next, we’ll examine California, a traditionally wealthy state with enough counties for reasonable examination. Similarly to the Bureau of Labor Statistics, I divided the counties in the US into quintiles, and visualized California by color – darker being more wealthy by adjusted salary. So let’s compare the maps. First is a simple zoom-in on California on the BLS map:

Screen Shot 2014-08-12 at 12.01.00 AM

This is a relatively dark map. Many counties near San Francisco, Los Angeles, and San Diego are the darkest color, or the highest quintile of weekly salary ($786 weekly and above). All the rest of the counties except for two of them are in the top 3 salary quintiles, and California has exactly zero counties in the bottom salary quintile ($587 weekly and below). The adjusted salary map tells a very different story:


A much more lightly-colored map (and poorly designed, but nobody said I was a graphic designer), by any stretch of the imagination. Well over half of the state falls into the bottom two quartiles (A score of 6.5 or less for adjusted salary). Only the usual suspects, the counties in and around Silicon Valley, along with San Diego county, are in the top quintile (7.7 or more for adjusted salary). Even Los Angeles county didn’t make the top quintile. It would seem the vast majority of California, though they bring in a high salary, are paying significant amounts of said salary to rent, groceries, transportation, healthcare, and the like.

So does this mean it’s not worth it to live in the five Californias that aren’t Silicon Valley? Don’t let me or anything said here tell you the answer to that. What I’ve come to learn, and what I hope most of us know is that numbers can be construed to say anything. This was a rather simple data study, but one that I hope makes us think about life in high-salary parts of the country. If you’re paying pennies on the dollar for rent and gas, don’t you have much more of your paycheck left over for leisure (which may also be less expensive in regions with a lower cost of living) even if your paycheck is less than someone doing the same job in New York?

I’d love to do more data studies like this in the future, so perhaps this won’t be strictly a Bitcoin blog as I move forward (expect plenty of Bitcoin talk anyway, maybe just not every single post). In addition, I’m nearly done designing my own website so I don’t have to live within the constraints of WordPress much longer. Expect big things in the next few weeks!

Thanks again to both the Bureau of Labor Statistics and The Council for Community and Economic Research for the use of their data!

Why Bitcoin is NOT like the Internet in 1994

This will be brief – not a whole lot to say, but I think it’s an important point to make. Jeremy Allaire, co-founder of Circle, has often asserted that Bitcoin is extremely similar to the Internet in 1994. I agree with him on a number of levels – it’s decentralized, it’s disruptive, and it has a lot of people inspired to do a lot of really cool things. However, Bitcoin’s revolution will not be like that of the Internet. Why? One simple reason: The Internet in 1994 did not have the Internet in 2014. Sounds silly, but if Internet circa 1994 could somehow have the communicative force behind it that Bitcoin does through Twitter and Facebook, the disruption wrought by the Internet would have taken a vastly different form. People didn’t know about the Internet when it was first invented, never mind anyone understanding it. This is simply not true with Bitcoin – anyone with any interest in finance or technology knows Bitcoin, at the very least. Bitcoin will spread faster than the Internet in 1994, due to none other than… the Internet.

While I have the utmost respect for Allaire and Circle, I can’t agree completely with his metaphor. Bitcoin will certainly have a level of disruption similar to or perhaps even greater than the Internet, but the disruptive paths the two will have followed will prove to not be similar at all.

Recent Patent Filings and Public Perspective

To preface this brief post, I’ll first say sorry for the little sabbatical I took – new season, new internship, it’s taken me a while to get into the swing of things. Secondly, I’ll throw you a pair of Coindesk articles detailing patents filed within the past few weeks by major companies with regards to Bitcoin. Both articles at least briefly discuss these patents as threats to the Bitcoin ecosystem – this is not the angle I wish to look at these patents from. I’m not a fan at all of patents restricting creativity and innovation in favor of maintaining the balance of power in business, and that’s exactly what these companies are trying to do, in my opinion. Are they a threat? The answer to that question is much less clear.

The very filing of these patents, though, is indicative of a major change in the view of Bitcoin in the realms of payments and ecommerce. Bitcoin has gone from being compared to the Quetzal, as mentioned in my recent paper, to being a force worthy of legal action. Legal action, by the way, specifically designed to secure the place of these finance companies in the business world as Bitcoin begins to invade their space. It’s not hard to see how Bitcoin could be a major threat to payments companies like MasterCard and Western Union, or even an ecommerce company like eBay. Yet it was largely ignored, at least publicly, by these very companies. Whether it’s the recent surge in the price of BTC back above the $600 mark or the press surrounding Bitcoin in light of events such as the Silk Road bitcoin auction, it would appear as if it’s becoming more and more difficult to ignore Bitcoin.

MasterCard wishes to add Bitcoin under the umbrella of its “global shopping cart”, a concept seeking to provide “flexibility in how a customer would fund a shopping cart purchase”, in the words of a MasterCard rep. Western Union seeks to have exclusive rights to an exchange allowing users to trade digital currency for fiat currency (which intrigues me, since that concept already exists rather prevalently). If granted, these patents not only protect these companies from complete disruption by Bitcoin, but suddenly make these companies major players in the realm of cryptocurrency – quite the 180, no?

Any supporter of cryptocurrency, or really anybody that’s even vaguely educated on the topic, must be saying “I told you so” to some extent as news like this continues to emerge. There’s certainly still a long battle for the crypto-community before anybody can really say Bitcoin is a mainstream concept, but as major finance companies begin to file faux-intellectual property claims out of what can only be described as fear, I think we can say we’re gaining ground. California legalized cryptocurrencies before it legalized other currently semi-legal substances. Should be absolutely fascinating to see how Bitcoin fares legally over the next few months.

A Study in Contrasts – Circle, QuickCoin, Armory

Hey all – life’s been crazy with finals and a graduation or two to attend. Glad to finally be back writing about Bitcoin. There’s been some exciting developments lately, but after attending the MIT Bitcoin Expo a few weeks back and learning of the release of a pair of ease of use-focused Bitcoin products (Boston-based Circle‘s product and QuickCoin), I want to speak on a philosophical level in this post.

Both Circle and a Bitcoin startup known as Armory had featured speakers at the MIT Bitcoin Expo. The products of these companies couldn’t be more different. While Armory is by no means difficult to use for most tech-literate people, it seemed to me that Armory placed little or no emphasis on being user-friendly. For example, a paper backup is the only way to back up and retrieve your wallet if you happen to forget a password. One also needs to install and synchronize the full Bitcoin-qt client and run a full node to use Armory online. Armory also has an option for an offline wallet for “cold storage” – possibly the most secure way to hold bitcoins. This requires an offline computer, though, which is simply not something every Bitcoin user is going to have on hand. (Read more about using Armory’s wallet here.)

Armory is still in a beta build, yet there have been no reported hacks or losses of bitcoins due to the software itself. If one loses their paper backup and something happens to the computer (online or offline) on which the wallet is hosted, though, bitcoins can be lost. Using Armory is inconvenient – no two ways about it. Installing and synchronizing Bitcoin-qt takes hours, and even after it’s installed, it can take a few minutes to load. An offline computer just seems completely unnecessary to own for any other purpose besides something like Armory. And paper is annoying. Yet security of a digital currency in a world full of hackers is an issue that may be paramount.

A common thread among the MIT expo speakers, though, was how Bitcoin would be eventually adopted by the masses. Circle co-founder Sean Neville was among the speakers, and raised some excellent points on the topic of Bitcoin adoption. To paraphrase, he more or less said that your everyday person isn’t going to use Bitcoin when an understanding of cryptography, the Block Chain, private/public keys, and other similarly complex ideas is needed. The way I understand it, Neville thinks that Bitcoin needs an app that “translates it into English”, so to speak. Enthusiasts have no trouble using and wrapping their heads around Bitcoin, but that’s simply not enough for the mainstream to adopt cryptocurrency technology.

Circle and QuickCoin, both of which launched early builds of their wallets recently, are focused on adoption and ease of use. Not to say these wallets are not secure – there’s no evidence yet that they are not secure products for the purchase and storage of bitcoins – but when compared to a wallet like Armory’s, one simply can’t make an argument that QuickCoin’s or Circle’s wallet is more secure. Yet both Circle and QuickCoin are drawing rave reviews for their ease of use (and lack of transaction fees, in Circle’s case).

These apps and other similar apps beg the question: What’s more important, security or adoption? A lack of security in a wallet will eventually give way to hackers stealing bitcoins. However, what’s the point of securing your bitcoins if neither none your peers nor your preferred merchants use bitcoins? If I had to choose, I’d side with adoption. The Internet had its issues with security as it grew, but these proved to simply be growing pains as the Internet gradually dominated the world.

Ultimately, security and adoption are likely to go hand in hand. Increased security will help adoption as users begin to view Bitcoin as safer, and a greater network of users will spur the need for greater security. Bitcoin certainly does not have a “killer app” yet that combines security and ease of use, but I believe one or more of these will emerge in the near future (perhaps even in the form of one of the aforementioned wallets). There isn’t only one credit card company in existence, and I don’t believe there will be only one Bitcoin company in the future. However, the companies that are able to harness the best of the seemingly opposite concepts of security and ease of use will be the winners when the dust clears.

You can check out the apps discussed in this article via the links below:




Innovation Within the Innovation – Ethereum

Bitcoin itself is a revolutionary technology. However, it will be how empowered individuals and startups innovate with it that truly defines how Bitcoin changes the world. “Innovation within the Innovation” examines people and companies using Bitcoin in ways that truly change the game.

The Bitcoin Expo in Canada was this past weekend, and it featured a hackathon in a language not many of us are familiar with. This turing-complete language and programming is called Ethereum – and the easiest way to describe it is programmable money. The Bitcoin community is excited across the board about the possibility of programming one’s own digital financial systems, and this excitement was on display at the Bitcoin Expo.

The concept of Ethereum is generally mind-blowing, but from what I can gather it’s based on the idea of being able to program digital “contracts” based in Ether, the currency of Ethereum. Once a contract is programmed, one can send a command or a “transaction” to the contract, causing the code within the contract to run. This could obviously cause any number of things to happen, from Ether being released to a given party, or any other kind of financial transaction occurring. This is obviously not particularly easy to wrap one’s head around. Cryptonomics mentions a few examples in this blog post: Savings accounts, peer to peer gambling, and even new currencies within Ethereum are more than possible.

Personally I can’t help but think of the idea of financial services becoming obsolete. Taxes could become automated. Financial contracts and remittance could become incredibly simple. I doubt humans will ever become obsolete in the world of finance, but a lot of cumbersome third parties have the potential to be eliminated if a platform like Ethereum catches on. It would seem that yet another sector of the workforce is destined to be replaced by computer scientists – surprise, surprise.

We first knew Bitcoin as a digital currency. A cryptocurrency, if you will. But now it would seem it is far more than that – many are calling this developing wave of further applications of Bitcoin the Internet of Money. For all the talk in the tech community about the Internet of Things, I think it’s painfully obvious which of the two is slightly gimmicky and which may completely change the way the world works. Tax day becoming an automated process might be a generation or more away, but Ethereum is more revolutionary an idea in finance than anything I’ve seen in a long time. I wish them the best.