GasPredictor.com
Home
  • Our latest prediction.

  • Prediction for Atlanta.
  • Prediction for Boston.
  • Prediction for Chicago.
  • Prediction for Denver.
  • Prediction for Houston.
  • Prediction for Kansas City.
  • Prediction for Las Vegas.
  • Prediction for Long Island.
  • Prediction for Los Angeles.
  • Prediction for Nashua, NH.
  • Prediction for New York.
  • Prediction for Oakland.
  • Prediction for Orlando.
  • Prediction for Phoenix.
  • Prediction for Portland, OR.
  • Prediction for Raleigh.
  • Prediction for St. Louis.
  • Prediction for San Diego.
  • Prediction for Seattle.
  • Prediction for Tacoma.
  • Prediction for Tampa.
    •
    Subscribe
  • Tomorrow's gas prices in your in-box every business day.

  • Subscribe for Atlanta.
  • Subscribe for Boston.
  • Subscribe for Chicago.
  • Subscribe for Denver.
  • Subscribe for Houston.
  • Subscribe for Kansas City.
  • Subscribe for Las Vegas.
  • Subscribe for Long Island.
  • Subscribe for Los Angeles.
  • Subscribe for Nashua, NH.
  • Subscribe for New York.
  • Subscribe for Oakland.
  • Subscribe for Orlando.
  • Subscribe for Phoenix.
  • Subscribe for Portland, OR.
  • Subscribe for Raleigh.
  • Subscribe for St. Louis.
  • Subscribe for San Diego.
  • Subscribe for Seattle.
  • Subscribe for Tacoma.
  • Subscribe for Tampa.
    •
    How to Use Gas Predictor
  • Exactly what do these predictions mean?

    •  Track Record
  • Over 99% correct since November, 2008

    •  Local Gas Prices
  • Current prices in various U.S. cities.

    •  Gas Saving Tips
  • Information you can use.
  • Gas Saving Articles.
  • Gas Saving Software.

    •  About Gas Predictor
  • Privacy Policy
  • We don't sell or otherwise abuse your information.

  • Contact Us

    •

    Announcing Our First Anniversary of Successful Gas Price Preditions!

    On this page:

    Summary of Results

    • General
      • We now publish 13 editions of the "Gas Predictor Newsletter," the National Edition and 12 local editions.
      • Nine of our editions have been correct 100% of the time, and four editions have had one incorrect prediction each.
    • National Edition, published since November 18, 2008, could hypothetically save you $10.72 per year.
    • Atlanta Edition, published since February 17, 2009, could hypothetically save you $11.24 per year.
    • Chicago Edition, published since March 9, 2009, could hypothetically save you $8.75 per year.
    • Denver Edition, published since October 26, 2009, could hypothetically save you $3.49 per year.
    • Houston Edition, published since March 9, 2009, could hypothetically save you $10.01 per year.
    • Long Island Edition, published since August 3, 2009, could hypothetically save you $13.00 per year.
    • Nashua, NH Edition, published since January 12, 2009, could hypothetically save you $8.43 per year.
    • Oakland Edition, published since June 8, 2009, could hypothetically save you $9.53 per year.
    • Raleigh Edition, published since February 17, 2009, could hypothetically save you $8.15 per year.
    • San Diego Edition, published since July 6, 2009, could hypothetically save you $8.54 per year.
    • Seattle Edition, published since September 21, 2009, could hypothetically save you $4.75 per year.
    • St. Louis Edition, published since October 12, 2009, could hypothetically save you $29.76 per year. NO, this is not a realistic claim, and we know it. See our explanation of why that savings is so far out of line with the other cities.
    • Tampa Edition, published since June 8, 2009, could hypothetically save you $7.77 per year.

    Our Hypothetical Test Case

    In the real world, of course, nobody can follow our buy/no-buy recommendations exactly. If you need a tank of gas every three days, and our prediction algorithm calls for declining gas prices five days in a row, you're going to have to buy gas on a day when we recommend waiting for a better price tomorrow. And if we predict rising gas prices tomorrow, and therefore recommend that you buy gas today, you're not likely to do it if you just filled up yesterday.

    So how can we measure the money saving power of our predictions?

    Of course, we have been following our own recommendations, as far as possible, since long before we began publishing our predictions. Why not just report our own savings? Well, each member of our team has radically different driving and fuel consumption habits. One of us drives a little econo-box and another drives a monster 4x4 pick-up that eats econo-boxes for breakfast. And by coincidence, every one of us has changed jobs within the past year (some more than once), so our driving and fuel requirements have all changed in the past year.

    We don't have a good, consistent, real-world result that covers the past year. So, we made one up. Not that we made up the results, just that we made up an artificial consistent gas-purchasing habit and "back tested" against our recommendations and the subsequent results.

    For each of our 13 published editions, we used the same hypothetical scenario. The hypothetical driver has a 14 gallon tank, and he uses exactly 2 gallons of gas each business day. He began following our recommendations on the day we began publishing each edition of the Gas Predictor. In each test case, he begins with a full tank in the morning.

    • At the end of the day, when he gets his Gas Predictor newsletter, if we recommend buying, he fills his tank, even if he can only buy two gallons.
    • If we recommend not buying gas, he does not buy gas unless his tank is empty. In that case, he buys only 4 gallons, to get him through a couple of days and allow himself the capacity to take advantage of a buying opportunity in the next couple of days.

    For each edition of the Gas Predictor Newsletter, we also ran a control case. The hypothetical control driver has a 14 gallon tank and uses exactly 2 gallons of gas each business day, just as in the test case. However, the control driver buys a full tank of gas whenever his tank gets down to 2 gallons.

    The bottom line, the numbers we claim above as the savings for each of our 13 editions of the Gas Predictor Newsletter, is the difference between how much the control driver would have spent on gas minus how much the test driver would have spent on gas. We also account for the fact that one or the other drivers will have more gas in his tank on the last day of our scenario (November 18, 2009). If one driver has more gas than the other, we calculate the value of that gas based on the retail price of gas on November 18, and subtract that from the amount spent on gas.

    In each city for which we publish a local Gas Predictor Newsletter, we use the retail price of gas as we recorded it on each day we have been publishing. That is, the second-lowest price of regular unleaded gas at about 9:00 AM local time in each respective city. For the test of the National Gas Predictor Newsletter, we used the average of those second lowest prices in each of the twelve cities that we use in our forecasting algorithm.

    back to top

    Why Can't We Save Money in Denver?

    And why do we appear to be generating literally incredible savings in St. Louis?

    These are two manifestations of the same underlying fact: We don't have enough of a track record to draw realistic extrapolations from.

    We have similarly unrealistic "results" in Denver, Seattle, and St. Louis. For similar reasons, our results in Long Island and San Diego should also be considered suspect even though, coincidentally, they are nearly in line with our more typical results for cities for which we have a longer track record.

    We observed and reported this phenomenon when we published a similar summary on the occasion of our 100th prediction, on April 3, 2009. At that time, running exactly the same hypothetical test scenarios on the 6 editions we were publishing then, we got unrealistically good results for Houston and unrealistically bad results for Chicago, each of which we had been predicting for less than a month. We stated that we were sure that, given more time, the results would become more realistic and more consistent. This has been borne out perfectly, as our results for Chicago and Houston are now very much in line with the results from our other cities. (Chicago is a little on the low side, but not too bad.)

    Along these lines, it is very interesting to note what happened to our "National Edition" forecasts since April. We stated at that time that our national predictions were the most realistic and reliable, because we had the longest track record with the national editon and because the national forecast begins with an average figure and therefore automatically smooths out the occasional blips in the data. In April, we extrapolated our annual savings from the national edition at $10.65 per year. Now, with an actual year's worth of data behind us, the savings works out to $10.72 per year. That's a bullseye, folks!

    We can repeat what we said more than half a year ago, now with enough real data to back it up: "We are supremely confident that as time goes by," our data will even out and the money-saving opportunities in our gas price predictions will be consistent.

    A realistic figure for saving money by using our predictions is somewhere between nine and thirteen dollars per year, centered right around eleven dollars, for someone who uses about ten gallons of gas per week.

    back to top

    What We Have Learned in This Analysis

    As we observed on the occasion of our 100th prediction, this first anniversary has given us an opportunity to take a backward look at what we are doing. Ordinarily, we only look at today's economic conditions and crunch the numbers to come up with tomorrow's prediction, with a quick look back to see that yesterday's prediction was correct. Today, we can take a good look over the past year.

    Generally, we are well pleased with what we have accomplished. We see three major areas for improvement and expansion, and we are working on all three.

    First, We Need to Account for Price War Conditions

    This was a lesson we learned with a few of our incorrect forecasts. We have learned how to identify regional price wars - when prices near one of "our" cities are moving downward without reference to movements in the broader commodities markets - but we have not yet developed a fully satisfactory way to account for the local pressure this creates. The problem is enormously complex, involving such factors as the price differential between affected and unaffected areas, the size of the affected area, and the distance between affected and unaffected areas. So far, we have not missed a price war-induced decrease in prices since we first began paying real attention to the phenomenon, but we don't believe this part of our algorithm is as reliable as it ought to be.

    We're working on it. Unfortunately, we believe we will need more data, which means that we will probably need more incorrect predictions before we know that we have the issue solved.

    Second, We Should Expand to Other Cities

    Our plans and goals are growing. Originally, this was just an idea to save ourselves some money on gas (back when it was hovering around four dollars a gallon). We did that.

    Then we "went public," and our goal was to refine our forecasting algorithm to fit the 12 cities we were using in our model. We did that. (Yes, it still needs more tweaking in the price war department, but what we have is pretty darn good.)

    Next, we thought we'd expand by applying our algorithm to other cities across the U.S. We solicited input from our Web site visitors, and got a few good suggestions. We set to work on it, and immediately ran into a problem.

    The problem's name is Ohio.

    We don't quite know why, but our forecasting algorithm doesn't work well at all for any city in Ohio. And several cities in Ohio are among the most requested cities for which our Web site visitors want us to publish gas price predictions.

    We think the problem with Ohio is related to the price war phenomenon, and to the fact that so many sizable cities are so close to each other in Ohio. Local market pressures, unrelated to wholesale gas prices, drive prices up and down wildly, and waves of price increases and decreases sweep back and forth across Ohio.

    By way of illustration, as this is being written (November 18, 2009), we have recorded the largest one-day price increase in any city we have ever monitored: 28 cents per gallon! And it happened in two separate cities on the same day! And, surprise! Both of those cities are in Ohio. (Columbus and Cincinnati, in case you were wondering.)

    Frankly, we're beginning to think that we've spent too much energy on Ohio. We've decided to stick it out until the end of the year, to figure out both the price war issue and the Ohio issue (which, as was stated, we think may be the same issue). Then we'll either publish a reliable daily forecast of gas prices in one or more cities in Ohio, or we'll move on to other regions.

    Third, We Should Predict Diesel Fuel Prices

    We already own several domain names related to predicting diesel fuel prices, and we know that there is some market interest in our plans. Unfortunately, after some initially promising tests, we have found that our forecasting algorithm does not work as well for diesel fuel as it does for gasoline.

    We're working on it. And we believe that the remaining problems are not nearly so complicated as the price war phenomenon. (But, of course, there is a price war phenomenon at work in diesel fuel pricing, just as there is for gasoline.) We are confident that we can develop a forecasting algorithm for diesel fuel that is just as reliable as our present algorithm for forecasting gasoline prices. It will only take a little more time. Not much.

    Join us for the adventure! Subscribe today!

    back to top
    StumbleUpon Toolbar Stumble It!
    Save This Page
    in del.ico.us

    [ Today's Gas Price Prediction ] [ Subscribe ] [ How to Use Gas Predictor ] [ Gas Predictor.com's Track Record ] [ Local Gas Prices ] [ Gas Saving Tips ] [ About Gas Predictor ] [ Privacy Policy ] [ Contact Gas Predictor ]