<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Ola Lidmark Eriksson on Medium]]></title>
        <description><![CDATA[Stories by Ola Lidmark Eriksson on Medium]]></description>
        <link>https://medium.com/@olalidmarkeriksson?source=rss-763ed10a10cd------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*XuyXqawXpB9DLCdi.jpg</url>
            <title>Stories by Ola Lidmark Eriksson on Medium</title>
            <link>https://medium.com/@olalidmarkeriksson?source=rss-763ed10a10cd------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Fri, 20 Nov 2026 02:19:20 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@olalidmarkeriksson/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Statistical Models what do they know? Do they know things?? Lets find out! #1]]></title>
            <link>https://medium.com/@olalidmarkeriksson/statistical-models-what-do-they-know-do-they-know-things-lets-find-out-1-c09443e0ddd6?source=rss-763ed10a10cd------2</link>
            <guid isPermaLink="false">https://medium.com/p/c09443e0ddd6</guid>
            <category><![CDATA[uncategorized]]></category>
            <dc:creator><![CDATA[Ola Lidmark Eriksson]]></dc:creator>
            <pubDate>Sat, 01 Apr 2017 14:43:12 GMT</pubDate>
            <atom:updated>2017-04-01T14:43:12.334Z</atom:updated>
            <cc:license>http://creativecommons.org/licenses/by/4.0/</cc:license>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/400/0*SThuo5CmQZlZS8_m." /></figure><p>Originally posted on <a href="http://ift.tt/2mYHexr">Cognitive Football</a>:<br> It’s a question as old as time, or at least as old as public football analytics. How good are our prediction models? To try and answer this question, I’ll be using the Ranked Probability Score (RPS). The RPS is a statistical method that measures the quality of predictions. It calculates…</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1/0*YxDBpqF1kLlTQU9j." /></figure><p><em>Originally published on </em><a href="http://ift.tt/2oKAW0G"><em>Wordpress</em></a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c09443e0ddd6" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[My bet bot]]></title>
            <link>https://medium.com/chatbot-news-daily/my-bet-bot-99ec2408da7a?source=rss-763ed10a10cd------2</link>
            <guid isPermaLink="false">https://medium.com/p/99ec2408da7a</guid>
            <category><![CDATA[football]]></category>
            <category><![CDATA[sports]]></category>
            <category><![CDATA[betting]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[ai]]></category>
            <dc:creator><![CDATA[Ola Lidmark Eriksson]]></dc:creator>
            <pubDate>Mon, 23 Jan 2017 23:40:32 GMT</pubDate>
            <atom:updated>2017-01-24T09:58:47.819Z</atom:updated>
            <content:encoded><![CDATA[<p>This is the story of how I outsmarted betting journalists with my AI based betting platform. It is a follow up to my <a href="https://medium.com/m/global-identity?redirectUrl=https://chatbotnewsdaily.com/what-i-learned-about-big-data-and-machine-learning-from-trying-to-predict-football-matches-2f81d019bea0">last post</a>, which also by the way, has become subject to a <a href="https://www.youtube.com/watch?v=3A4EBiNrrI0">TedX talk</a>. But above all this is my manifesto about how I think betting advice should be done.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2F3A4EBiNrrI0%3Ffeature%3Doembed&amp;url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3D3A4EBiNrrI0&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2F3A4EBiNrrI0%2Fhqdefault.jpg&amp;key=d04bfffea46d4aeda930ec88cc64b87c&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/2b9788cfe1f2bfc3967530d4cf6c5462/href">https://medium.com/media/2b9788cfe1f2bfc3967530d4cf6c5462/href</a></iframe><figure><img alt="" src="https://cdn-images-1.medium.com/max/400/1*Qzc0fqkDTZVpZAm6eJ726g.png" /></figure><h3>Betbot.soccer</h3><p>First of all I want to give you the result of four years work — <a href="http://www.betbot.soccer/">betbot.soccer</a><br>Betbot is my attempt to showcase the strength of my model which predicts the outcome of football games. I have tried to visualize everything in a way that anyone and everyone could understand. Since I wrote the <a href="https://medium.com/m/global-identity?redirectUrl=https://chatbotnewsdaily.com/what-i-learned-about-big-data-and-machine-learning-from-trying-to-predict-football-matches-2f81d019bea0">last Medium article regarding my work</a> I have made some modifications to the model as such, resulting in that it now actually predicts all football games with at least <strong>50%</strong> accuracy. Now, that may not sound much but keep in mind that every game has three possible outcomes. The platform also strives to make the predictions understandable and intuitive by making it possible to filter out games that are more or less probable to end as the model predicted. With the highest indications of a certain outcome, what I call “extremely likely” — Betbot is <strong>80% </strong>accurate.</p><h3>Why???</h3><p>Now you might ask yourself why spend so much at this project. Especially since it more or less seems to be a dead end when it comes to making money. For me the answer has always been <em>curiosity</em>. Curiosity and an eagerness to <em>learn by doing</em>. During my years as a developer there is nothing I ever have done that has taught me more than this project. So for me the main goal was never to create something that would make me rich or make others poor. If I now, after four years of work would make some money out of it — that would be a bonus. And the money made would probably not come from betting.</p><p>I think this is where my approach really differs from all the other similar projects and sites out there. And now I’d like to say something about them — not for being patronising. I just want to explain how my approach differs and in my opinion is more <em>fair</em>.</p><h3>The lack of why</h3><p>I guess most of you reading this already have seen <a href="https://www.youtube.com/watch?v=IPYeCltXpxw">Simon Sineks legendary Ted talk about <em>purpose</em></a>. The lack I see in the betting advice industry is just that — a purpose — a true <em>what</em>, other than making money.</p><p>The whole idea of the betting industry is that in order for someone to make money, someone else has to lose money. And to have a business model that contradicts this makes me confused. I mean — if everybody were using betting advice no one would ever lose money, in theory. That really contradicts the very foundation of the entire betting industry.</p><p>There are a lot of services out there offering this type of services. Here are a few of them:</p><ul><li><a href="http://www.dynamobetting.com/">DynamoBetting Home page - Get your Free Trial then get on Winning Tips and join the Number 1 Tipster Group</a></li><li><a href="https://www.betterbet.bet/">betterbet.bet</a></li><li><a href="https://pyckio.com/">Looking for the best Tipsters in the world</a></li></ul><p>You might notice that a common denominator of the examples above is that none of them actually reveals how good they perform — how often their predictions and advice actually is right.</p><p>There are sites/platforms that does tell you how well their models work. Some of them are:</p><p><a href="http://www.stratabet.com/">https://www.stratabet.com/</a></p><p><a href="http://www.clubgowi.com/">https://www.clubgowi.com/</a></p><p><a href="https://www.johnnybet.com/app/picks#stats?cf%5Bday_tip%5D=1">Betting Predictions, Tips, Picks, Odds, Previews - JohnnyBet</a></p><p>Note that Johnnybet actually has a negative ROI, return of investment if you filter out only football bets.</p><p>Still — the above looks good. They seem to make money right? Well there is another problem that is well described by <a href="https://twitter.com/12Xpert">Joseph Buchdal</a> in <a href="https://www.pinnacle.com/en/betting-articles/betting-strategy/the-law-of-small-numbers-in-sports-betting">this article</a>. In short it describes a classic problem called the <em>Hospital Quiz</em> that basically says that looking at patterns in small amounts of samples really doesn’t say anything about long term patterns. So for me the fact that Stratabet has proved to be successful over 90 bets really doesn’t say anything about how successful they will be after 900 bets or 9000 bets.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*X2n1vs7RBSW7vUcYoTYjrA.png" /></figure><p>Actually, my model seems to perform just as well as Stratabet if I filter out only games with a high likelihood to end as predicted in combination with relatively high odds. But I would not dare to say that this in any way is data enough to prove that I have a model that can make money over time.</p><blockquote>And for the few that actually are candid with how their betting performs — to all of you using them — I am sad to say that buying stocks in a random betting company would have been a better way to make money.</blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/760/1*N9FiNpLgMOg-adxPQk3m8g.png" /></figure><p>Look for instance at the development of Betfairs stock the last five years. When any betting advisory platform can show the same constant development over five years time — I will rest my case.</p><h3>How</h3><p>I don’t know of anyone else that is totally open about what they do and what results they have (Stratabet excluded). For me — providing the results is the whole aim. This is also why Betbot.soccer as far as I know, is the only model online that actually displays the accuracy of the predictions, not only in ROI, but also in actual numbers. For me it has always been the actual number of correctly predicted games that has been most interesting. It is also the one thing that has kept me going. The question always being how many games you can predict correctly. Over the past six months my predictions has competed with the predictions of experts and newspapers in Sweden in a game called Stryktipset. The idea is to correctly predict 13 games. My model actually has outperformed experts. See for yourself at <a href="http://tipper.se/">http://tipper.se/</a>!</p><h3>What</h3><p>When I made Betbot the key words was <em>simplicity</em> and <em>transparency</em>. Anyone is supposed to understand how my model works and how well it performs. It is first and foremost NOT a betting advice site but an AI powered game preview generator. Therefore there is a <a href="http://www.betbot.soccer/stats">stats</a> section where you instantly can see how well the model has performed historically during different time periods and with different strengths in probabilities. And there are actual numbers there to directly give you an indication of how many games that are correctly predicted in relation to incorrectly predicted. Yes there are some ads on the site but that is only an attempt to earn enough to pay for the server power needed to have the site up and running.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/412/1*MP3HYEAWesQVlV-MMOsDHg.png" /></figure><p>And since the project for me is about making the best possible <em>unbiased</em> guesses about outcome rather than making the best bets in order to make money, I also put down an effort of trying to visualize some of the key statistics behind the predictions.</p><p>The idea behind the site is then to filter out games that would be extra interesting to bet on for instance (sorry I haven’t found out any better use yet). The basic idea would be to start looking for games that has a high probability of a certain outcome with relatively high odds. That can be achieved by setting the filter to “Likely” or “Very likely” in combination with “High odds”. An example of a game that then can be found is here to the left.</p><p>What I also did was that I used all the data I have to generate characteristics of every upcoming game and create texts about them. But that is a whole other story…</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=99ec2408da7a" width="1" height="1" alt=""><hr><p><a href="https://medium.com/chatbot-news-daily/my-bet-bot-99ec2408da7a">My bet bot</a> was originally published in <a href="https://medium.com/chatbot-news-daily">Chatbot News Daily</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Damallsvenskan summary and The Mark Dempsey effect]]></title>
            <link>https://medium.com/@olalidmarkeriksson/damallsvenskan-summary-and-the-mark-dempsey-effect-4a3b19195c8b?source=rss-763ed10a10cd------2</link>
            <guid isPermaLink="false">https://medium.com/p/4a3b19195c8b</guid>
            <category><![CDATA[season-summaries]]></category>
            <category><![CDATA[djurgårdens-if]]></category>
            <category><![CDATA[damallsvenskan]]></category>
            <category><![CDATA[expected-goals]]></category>
            <category><![CDATA[player-analysis]]></category>
            <dc:creator><![CDATA[Ola Lidmark Eriksson]]></dc:creator>
            <pubDate>Fri, 18 Nov 2016 11:29:08 GMT</pubDate>
            <atom:updated>2016-11-18T11:29:08.540Z</atom:updated>
            <cc:license>http://creativecommons.org/licenses/by/4.0/</cc:license>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/400/0*KwAlm8S4jk5py6JH." /></figure><p>Just a brief summary of Damallsvenskan. Not at all as extensive as I did last year, but still. Teams A plot of the teams performance with actual goals as Y coordinate and expected as X really says that most teams at the end of season didn´t have much of under/over performance when it came to … <a href="http://ift.tt/2faMubw">Continue reading Damallsvenskan summary and The Mark Dempsey effect</a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1/0*6m5g1Q8TAEOVVfYO." /></figure><p><em>Originally published on </em><a href="http://ift.tt/2faMubw"><em>Wordpress</em></a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=4a3b19195c8b" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The story of GIF Sundsvall]]></title>
            <link>https://medium.com/@olalidmarkeriksson/the-story-of-gif-sundsvall-c12eb568facc?source=rss-763ed10a10cd------2</link>
            <guid isPermaLink="false">https://medium.com/p/c12eb568facc</guid>
            <category><![CDATA[season-summaries]]></category>
            <category><![CDATA[expg]]></category>
            <category><![CDATA[expected-goals]]></category>
            <category><![CDATA[projection]]></category>
            <category><![CDATA[allsvenskan]]></category>
            <dc:creator><![CDATA[Ola Lidmark Eriksson]]></dc:creator>
            <pubDate>Sun, 13 Nov 2016 17:59:21 GMT</pubDate>
            <atom:updated>2016-11-13T17:59:21.748Z</atom:updated>
            <cc:license>http://creativecommons.org/licenses/by/4.0/</cc:license>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/400/0*yeEzyVZ2QKumtpIm." /></figure><p>This season I won´t do a regular season summary of Allsvenskan the way I did last year. Instead I will tell the story about GIF Sundsvall, and introduce a new area chart visualization Giffarna started the season really well, winning lots of close games and actually also at one point, leading Allsvenskan. Their underlying numbers told another tale, … <a href="http://ift.tt/2fwpqCf">Continue reading The story of GIF Sundsvall</a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1/0*UGyWFYGhBp1sTNGy." /></figure><p><em>Originally published on </em><a href="http://ift.tt/2fwpqCf"><em>Wordpress</em></a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c12eb568facc" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The right footed left forward]]></title>
            <link>https://medium.com/@olalidmarkeriksson/the-right-footed-left-forward-5a9e188c6547?source=rss-763ed10a10cd------2</link>
            <guid isPermaLink="false">https://medium.com/p/5a9e188c6547</guid>
            <category><![CDATA[statiscics]]></category>
            <category><![CDATA[expected-goals]]></category>
            <category><![CDATA[expg]]></category>
            <dc:creator><![CDATA[Ola Lidmark Eriksson]]></dc:creator>
            <pubDate>Fri, 21 Oct 2016 12:37:27 GMT</pubDate>
            <atom:updated>2016-10-21T12:37:27.589Z</atom:updated>
            <cc:license>http://creativecommons.org/licenses/by/4.0/</cc:license>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/400/0*3D7lJ0_G-eSl5d-7." /></figure><p>Does left footed football players score more goals from the right? And do right footed players score more from the left?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1/0*d6XY1iiI8leUNtTC." /></figure><p><em>Originally published on </em><a href="http://ift.tt/2duPXxP"><em>Wordpress</em></a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5a9e188c6547" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Allsvenskan 2016 some early stats]]></title>
            <link>https://medium.com/@olalidmarkeriksson/allsvenskan-2016-some-early-stats-29cdc08041d2?source=rss-763ed10a10cd------2</link>
            <guid isPermaLink="false">https://medium.com/p/29cdc08041d2</guid>
            <category><![CDATA[statiscics]]></category>
            <category><![CDATA[expg]]></category>
            <category><![CDATA[projection]]></category>
            <dc:creator><![CDATA[Ola Lidmark Eriksson]]></dc:creator>
            <pubDate>Tue, 26 Apr 2016 17:54:55 GMT</pubDate>
            <atom:updated>2016-04-26T17:54:55.053Z</atom:updated>
            <cc:license>http://creativecommons.org/licenses/by/4.0/</cc:license>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/proxy/0*XINwanwZO-KSRwTl." /></figure><p>Since we have two really unexpected teams in the first two spots of Allsvenskan after 5 rounds I thought it would be interesting to look for some explanations in the numbers. I also realized when the blog post was nearly done that it may serve as a brief explanation of XG-G on a team and … <a href="http://ift.tt/24i0CRO">Continue reading Allsvenskan 2016 — some early stats</a></p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/0*MSuGbMKcW2W82Ixe." /></figure><p><em>Originally published on </em><a href="http://ift.tt/24i0CRO"><em>Wordpress</em></a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=29cdc08041d2" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Update]]></title>
            <link>https://medium.com/@olalidmarkeriksson/update-5e86b8981b19?source=rss-763ed10a10cd------2</link>
            <guid isPermaLink="false">https://medium.com/p/5e86b8981b19</guid>
            <category><![CDATA[league-tables]]></category>
            <category><![CDATA[medium]]></category>
            <category><![CDATA[projection]]></category>
            <dc:creator><![CDATA[Ola Lidmark Eriksson]]></dc:creator>
            <pubDate>Thu, 24 Mar 2016 15:35:44 GMT</pubDate>
            <atom:updated>2016-03-24T15:35:44.764Z</atom:updated>
            <cc:license>http://creativecommons.org/licenses/by/4.0/</cc:license>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/proxy/0*xRK5gsR3jU8vPZLd." /></figure><p>I just wanted to post a brief update of what I have been working on lately. The Swedish leagues are soon starting and the work I’m doing together with ÖFK is making some real progress. So expect a lot more on the blog in the coming months! Last week I wrote a blog post about … <a href="http://ift.tt/1pAcoHW">Continue reading Update</a></p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/0*f3vDRrJJD6w0Li88." /></figure><p><em>Originally published on </em><a href="http://ift.tt/1pAcoHW"><em>Wordpress</em></a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5e86b8981b19" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[What I learned about Big Data and Machine Learning from trying to predict football matches.]]></title>
            <link>https://medium.com/chatbot-news-daily/what-i-learned-about-big-data-and-machine-learning-from-trying-to-predict-football-matches-2f81d019bea0?source=rss-763ed10a10cd------2</link>
            <guid isPermaLink="false">https://medium.com/p/2f81d019bea0</guid>
            <category><![CDATA[chatbots]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[big-data]]></category>
            <category><![CDATA[tech]]></category>
            <category><![CDATA[bots]]></category>
            <dc:creator><![CDATA[Ola Lidmark Eriksson]]></dc:creator>
            <pubDate>Fri, 18 Mar 2016 15:28:38 GMT</pubDate>
            <atom:updated>2016-12-21T07:50:18.117Z</atom:updated>
            <content:encoded><![CDATA[<p>Two years ago I asked myself if it in any way would be possible to use Machine Learning techniques to predict the outcome of football matches.</p><h3>Data mining</h3><p>To describe the process briefly I started by collecting as much data as I could get hold of. I mined data about old games from every different source and API I could find. Some of the more important ones were Football-data, Everysport and Betfair. I then took all the data for from the old matches, with its corresponding results, quantified it and put it in a database. Finally I used the data to train a Machine Learning model, using it to predict upcoming games.</p><h3>How to measure how well a model performs</h3><blockquote>Now, the nature of a football game is, of course, that it is unpredictable. I guess that is why we love the game.</blockquote><p>Now, the nature of a football game is, of course, that it is unpredictable. I guess that is why we love the game. But I still was a bit obsessed by the naive idea that I with a Machine Learning approach was going to be able to predict games better than I would using my own mind. I knew from the past that I, as most humans do, base predictions on emotions, rather than facts and that I am somewhat biased. I know for myself that I quite often placed bets in the past based on a “gut feeling”.</p><p>The first question I now had to answer was how to measure if my Machine Learning model was successful or not. I quite quickly came to realize that measuring the actual percentage of correctly guessed games didn’t say that much if I didn’t put it in relation to something else. And the best thing I could come up with relating the model to was what other people were thinking. The easiest way to assess that would be to look at market regulated odds. So I started comparing how my model would perform if betting on Betfair because their odds are regulated on people betting against each other, making the odds a reflection of what the “market” predicts.</p><h3>The result</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*9L-pBgC4Da_k14cv." /></figure><p>So now, two years has passed. Has the model made me rich? No, not at all. Quite soon I realized that the predictions my model made for most part was aligned with the market. Since I use a regression based model I’m able to predict the strength of the probability of a certain outcome of a game. And at the strongest grades of probability my model gives, it predicts roughly 70% of the games correctly. Now the problem is that the market more or less performs just as well making it hard to actually make money out of my model. But, to be honest, I never really thought that I would create a money machine. Instead I have come to several insights about the possibilities and limitations of Big Data and Machine Learning.</p><h3>How much does a model learn over time?</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*lpgeKwlnZKvrP0Sn." /></figure><p>One of the first things I started looking at was that, since the nature of Machine Learning would be that it in theory gets better over time — as the amount of data the model has to learn from grows, the outcome of the predictions would improve. This is something I haven’t seen at all. Two years ago I started with having about 2000 games in my database with quite a limited dataset attached to them. Now I have almost 30000 games, complete with lots of data covering everything from weather and distances between the teams home grounds to shots and corners for and against. So, given all this data and the fact that the model has been able to “learn” over time still hasn’t improved the predictions. This has taught me that machine learning only takes you so far in trying to predict the unpredictable.</p><h3>Generalization</h3><p>Another important lesson I have learned is the power of machine learning still in many ways lies in its power to make unbiased generalizations. Over the past two years I was very curious to see if my model could predict when winning or losing streaks were to be broken. If it for instance could predict when Barcelona would finally lose after winning 10 straight games. If the model could find small signs that would indicate some kind of anomaly. Well, it has shown to not be that good at that.</p><p>What I found instead was that it was really good at, over time bet against over valued teams. Last season I for instance saw how my model quite often predicted against Borussia Dortmund while the market made another prediction. Dortmund ended up having a bad season making my model really successful here in relation to the market. This season I have seen the same when it comes to teams like Liverpool and Chelsea. So the lesson learned is that some people tend to make decisions based on emotions. Liverpool and Dortmund are teams liked by lots of people and at times you make predictions with your heart instead of your brain. My Machine Learning model does not.</p><h3>Easy gains</h3><p>Last but not least I guess I learned that making better predictions than the market is hard. Still, when I started looking at what I actually had achieved instead of looking at what I hadn’t I realized some quite amazing things. From a simple Python program and less than 10000 rows of code I still had made something that performed just as well as the market. How many man-hours aren’t behind bookies odds models and predictions? My model also is able to, on a weekly basis pick out interesting bets, just as any newspaper or expert does, but theirs with lots of manual labour behind. So the main insight is that by making generalizations you might not be able to find the one bet that makes you rich but it may save lots of time placed in the correct context.</p><h3>Wide Ideas</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*tbhvfImhH9xcuJMR." /><figcaption>Visualization of ideas based on different parameters. The outliers represent the most promising ones. In large dataset this makes the process of selecting good ideas over bad more efficient.</figcaption></figure><p>With these insights I started to look at another project I’ve been involved in for the last 5 years. The idea platform <a href="http://getwideideas.se/">Wide Ideas</a>. What I wanted to do was to look at the ideas companies gather from their employees and try to predict whether the idea would be implemented or not.</p><p>We started by looking at the ideas just as if it was a football match. We quantified the data but instead of shots and weather we looked at how many who had interacted with an idea and in what way. Due to discretion I won’t go into details but the outcome was quite similar to the football model previously described. We now can make a quite good prediction on whether an idea will be implemented or not given the data the idea contains. This is a way of generalizing the ideas, answering the question, in general, what are the factors behind a good idea?</p><p>However — can we find a good idea that doesn’t follow the general patterns of a successful idea? No, not really — not yet at least. Still, for the product, and given that you look at an organisation that creates say 10000 ideas per year, finding any good idea is really hard and time consuming. So just by going from 10000 ideas to 100 probably good ideas and visualizing the result saves a lot of time. And this is where Machine Learning has given us the most gain.</p><h3>Predicting the unpredictable</h3><p>To sum my thoughts up. We see companies gathering lots of data promising that they might be able to predict anything from finding cancer to making self driving cars. And they might. Especially where generalization saves time. The medical implementations I think is a good example of this. Looking at pictures of birthmarks a Machine Learning model can pick the most likely ones to be cancer from a large set of pictures saving doctors important time and money.</p><blockquote>In what way is human behaviour predictable? How far can we come in predicting the human behaviour if it essentially is unpredictable?</blockquote><p>But a lot of the things companies may try to predict has an unpredictable nature. Human behaviour is one. In what way is human behaviour predictable? How far can we come in predicting the human behaviour if it essentially is unpredictable? We will be able to generalize, placing people into different categories based on what you like to eat, watch or do, but honestly, who likes to be generalized?</p><p>What the past two years has taught me is that we in some way right now may be seeing a Big-Data-bubble. Will Big Data really find the anomalies or will it just be really good at making generalizations? I often believe that many of the promises made by companies tend to be that they will be able to find the needle in the haystack but that the results most often are generalizations. I guess that one of the reasons they do this is because their values as companies right now often are based on the amount of data they possess and not what they do with it. And if they were honest with the fact that the make generalizations, good ones but still generalizations, their value would decrease. I hope that we can see a future where companies values are based on what they do with the data rather than how much data they have. This will require transparency and honesty, just as I’ve been with my football model.</p><p>So, until someone proves me wrong I’m not convinced in the power of Big Data in general. I only believe in it where the cases are clear and one of the most obvious and best ones are examples within healthcare. The risk otherwise is that you end up with as much data that the sheer amount suffocates every possibility to make sense out of it in any other way than vast generalizations.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=2f81d019bea0" width="1" height="1" alt=""><hr><p><a href="https://medium.com/chatbot-news-daily/what-i-learned-about-big-data-and-machine-learning-from-trying-to-predict-football-matches-2f81d019bea0">What I learned about Big Data and Machine Learning from trying to predict football matches.</a> was originally published in <a href="https://medium.com/chatbot-news-daily">Chatbot News Daily</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[3 more projections]]></title>
            <link>https://medium.com/@olalidmarkeriksson/3-more-projections-fc157993675e?source=rss-763ed10a10cd------2</link>
            <guid isPermaLink="false">https://medium.com/p/fc157993675e</guid>
            <category><![CDATA[la-liga]]></category>
            <category><![CDATA[projection]]></category>
            <category><![CDATA[bundesliga]]></category>
            <dc:creator><![CDATA[Ola Lidmark Eriksson]]></dc:creator>
            <pubDate>Wed, 13 Jan 2016 16:09:15 GMT</pubDate>
            <atom:updated>2016-01-13T16:09:15.906Z</atom:updated>
            <cc:license>http://creativecommons.org/licenses/by/4.0/</cc:license>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/proxy/0*AMszCqeHfxdeqCYF.png" /></figure><p>While I was at it I made three more projections at more or less the half way mark of the seasons. First Bundesliga: No big upsets to await. Bayer Leverkusen seem to have the best possibility to reach the third place but that race is the most open one with at least 5 teams having a … <a href="http://ift.tt/1PrBP48">Continue reading 3 more projections</a></p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/0*leLh4wfuxyeBfhVC.gif" /></figure><p><em>Originally published on </em><a href="http://ift.tt/1PrBP48"><em>Wordpress</em></a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=fc157993675e" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>