<?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 Paul DeVos on Medium]]></title>
        <description><![CDATA[Stories by Paul DeVos on Medium]]></description>
        <link>https://medium.com/@PaulDeVos7?source=rss-608456fc7b77------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*icJ2DdccYHWcgIpa.jpg</url>
            <title>Stories by Paul DeVos on Medium</title>
            <link>https://medium.com/@PaulDeVos7?source=rss-608456fc7b77------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Thu, 19 Nov 2026 23:03:34 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@PaulDeVos7/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[Cloud Providers, Virtual Reality, and NVIDIA]]></title>
            <link>https://medium.com/@PaulDeVos7/cloud-providers-virtual-reality-and-nvidia-74f7b8226f28?source=rss-608456fc7b77------2</link>
            <guid isPermaLink="false">https://medium.com/p/74f7b8226f28</guid>
            <category><![CDATA[cloud-computing]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[nvidia]]></category>
            <dc:creator><![CDATA[Paul DeVos]]></dc:creator>
            <pubDate>Sun, 29 Oct 2017 13:21:52 GMT</pubDate>
            <atom:updated>2017-10-29T13:21:52.461Z</atom:updated>
            <content:encoded><![CDATA[<p>Definitely will be interesting to see more specialized options in the coming years (or even months!). We’ve seen similar trends in blockchain technologies (e.g. coin mining rigs) and database backends (SQL vs NoSQL) — that is, more specialized solutions for more narrow problems vs general solutions for specific problems.</p><p>NVIDIA really is an amazing story as is AMD. Especially as we near such changes in technology. Also my bets on VR come from the leaders in VR already, aka video game companies (Take 2, EA, etc).</p><p>How Machine Learning Will Disrupt The Established Cloud Providers via <a href="http://twitter.com/forbes">@forbes</a></p><p><a href="https://www.forbes.com/sites/forbestechcouncil/2017/10/24/how-machine-learning-will-disrupt-the-established-cloud-providers/#321e2a6e2015">https://www.forbes.com/sites/forbestechcouncil/2017/10/24/how-machine-learning-will-disrupt-the-established-cloud-providers/#321e2a6e2015</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=74f7b8226f28" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Your time is limited, so don’t waste it living someone else’s life — Steve Jobs]]></title>
            <link>https://medium.com/@PaulDeVos7/your-time-is-limited-so-dont-waste-it-living-someone-else-s-life-steve-jobs-4b169a6c0d0c?source=rss-608456fc7b77------2</link>
            <guid isPermaLink="false">https://medium.com/p/4b169a6c0d0c</guid>
            <category><![CDATA[work]]></category>
            <category><![CDATA[steve-jobs]]></category>
            <category><![CDATA[time]]></category>
            <category><![CDATA[life]]></category>
            <category><![CDATA[time-management]]></category>
            <dc:creator><![CDATA[Paul DeVos]]></dc:creator>
            <pubDate>Fri, 30 Jun 2017 13:12:31 GMT</pubDate>
            <atom:updated>2017-06-30T13:13:08.596Z</atom:updated>
            <content:encoded><![CDATA[<p><a href="https://www.ted.com/talks/steve_jobs_how_to_live_before_you_die">Steve Jobs: How to live before you die</a></p><p>This isn’t a “quit your job” post. Work, or a job, is most often the way in which you serve others by providing either services or goods from your unique skills and abilities to better the lives of others. Very few will serve others in their leisure.</p><p>Most of us never get to do what we really want to do because we deem it too far away. The road and journey to what you want is often non-trivial.</p><p>To get to that state of service where you provide value to others you’re going to have to be able to harvest time to get develop those skills or ideas that are useful to others. And that’s what I’m reminded of when I watch this video. Redeeming your time. Your time is limited.</p><p>Make sure you’re really spending your time in such a way that you’ll get to use your abilities well. Is what you want now in line with what you want in five years? Discipline is choosing between what you want now and what you want most. Redeem the moments. The 90 seconds waiting for the microwave for your food to cook. Pushups? Air squats? Stretching? Email mom? Re-read a quote that you are trying to incorporate into your life? These little things may seem small or even insignificant, but they add up over time. Make those moments count. Luck is when opportunity and hard work finally meet. Make sure you’re ready for the occasion if it arrives.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=4b169a6c0d0c" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[2017 NBA Draft: FG vs Assists (Warning: Graphic Material)]]></title>
            <link>https://medium.com/@PaulDeVos7/2017-nba-draft-fg-vs-assists-warning-graphic-material-24c051995d85?source=rss-608456fc7b77------2</link>
            <guid isPermaLink="false">https://medium.com/p/24c051995d85</guid>
            <category><![CDATA[nba-draft]]></category>
            <category><![CDATA[basketball-analytics]]></category>
            <category><![CDATA[sports]]></category>
            <category><![CDATA[2017-nba-draft]]></category>
            <category><![CDATA[matplotlib]]></category>
            <dc:creator><![CDATA[Paul DeVos]]></dc:creator>
            <pubDate>Tue, 16 May 2017 22:21:00 GMT</pubDate>
            <atom:updated>2017-05-16T22:29:26.334Z</atom:updated>
            <content:encoded><![CDATA[<p>The point guard arguably has the most affect on a team’s ball movement, especially seen in assists. I say arguably, as with some teams another player might have a larger share of the team’s usage (or loosely, possessions) and therefore dictates the ball movement of the team. Basically, it might be safest to say that the team creation abilities of the highest usage player dictates whether that team would be classified as having good basketball movement. There are no doubt subtractors to having a good ball movement team. The 1980’s Celtics would often joke that they’d avoid passing the ball to Kevin McHale because he was a “black hole”, in effect, they’d never get the ball back. Luckily for those Celtics teammates, McHale was very efficient at converting the opportunities he got. And as a whole the Celtics were very successful with McHale as <a href="http://www.basketball-reference.com/players/m/mchalke01/splits/">they won 69.4% (674 of 971) of the games McHale played in</a>. Basketball-Reference doesn’t have the exact +-/ for McHale, likely due to incomplete play-by-play logs, but would probably assume Celtics were better with him than not.</p><p>But the reason I bring up the point guard and this ‘black hole’ theory is because in watching a number of NCAA games this year I noticed some teams had players who were “open” a lot, but had teammates who did not get them the ball. The reasons can vary obviously, possibly from 1) simply lacking the court awareness, 2) having their own prerogatives (*cough* ball hog! *cough*) or 3) perhaps a more strict adherence to a play for a given play. I tend to believe in the former two especially when the shots would be high percentage shots — often backdoor opportunities for layups and dunks or open corner 3s. While I do enjoy the “new” NBA with it’s more wide open basketball, there is some collateral damage from “opening it up” as depicted by the graphic below:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/576/1*NaAaJprZ4zBs11jZvlyflA.jpeg" /></figure><p>The danger in the “analytics” or more so the <em>application</em> of the analytics is we lose sight of the forest for the trees. <em>You</em>, might not fit the minimum ability needed to make a 3pt shot a “good shot” for the team. Especially in regards to forgetting that the best shot value is still the layup.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/339/1*Oqbb-PbGiVoE3oEX8SLTAA.png" /><figcaption>Bottom 20 of the Top 50 SRS Ranked Teams for 2017</figcaption></figure><p>In watching parts of about 150 NCAA games this year I can’t recollect a team was as awful as the Florida State Seminoles at moving the ball as well as <em>missing</em> the open man. A look the graphic to the side would tell you my eye test wasn’t that far off. Among the top 50 teams according to Basketball-Reference’s SRS rating Florida State ranks 8th from the bottom. And what’s especially starting is you’re watching a team that is 10 deep while putting out three very fast wings and 2–3 frontline players measuring 6&#39;11 to 7&#39;4 at all times. Michael Ojo is 7&#39;1 315 lbs and looks he could play NFL tight end or bodyslam The Big Show on WWE with ease. But the player who was most open without teammates getting him the ball was <strong>Jonathan Isaac</strong>, a consensus top 10 prospect in the 2017 NBA Draft. The one major knock on him is he isn’t a creator or that smart offensively. And sure, if you look at the box score stats or even more advanced stats like Synergy stats you’re going still believe he’s not much of a creator. But the stat’s can’t tell missed opportunities. The stats can generalize it though and they’re less subjective. I would also note that very few people would say Isaac has poor awareness on defense, a form of basketball IQ. So looking into the offensive awareness again might be worth the time. Preferring to not distract you with posting 100 video clips of Florida State plays where FSU players missed open teammates, especially Isaac, lets just look at some stats.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1004/1*ymth_j4E9PAg5n6nxToAxA.png" /><figcaption>Image created with Matplotlib</figcaption></figure><p>The graphic clearly shows that Florida State was bad, although, not quite as bad as Duke. Maybe Jayson Tatum also has more court awareness than we give him credit for? In both cases the guards for either team were far from “past first” players. Looking at the top assist guard in the nation, Lonzo Ball, we can see that UCLA is head and shoulders above the rest of the teams in 2017 in rate, but also in quantity. Only UNC, just below them in rate, just ahead in field goals made (lower right blue dot to them) is close.</p><p>For fun I decided to show UCLA in three ways: 1) with Lonzo Ball, 2) without Lonzo Ball, and 3) with a replacement of Bryce Alford (the 2016 UCLA PG) in his place. Note, with Alford, they’re about average. Without Ball, UCLA is almost identical to Florida State and Duke.</p><p>Obviously this still doesn’t take into account more factors and more “event level” data (as apposed to yearly aggregates), but nonetheless it might beg the question about player awareness and how we might figure out [if there are] ways to better tell offensive awareness from NCAA boxscore data. Another question might be, who were players who were NOT major usage players, had low assists and labeled “low offensive awareness”</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=24c051995d85" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Why I’m starting this blog]]></title>
            <link>https://medium.com/@PaulDeVos7/why-im-starting-this-blog-983c1248b2c6?source=rss-608456fc7b77------2</link>
            <guid isPermaLink="false">https://medium.com/p/983c1248b2c6</guid>
            <category><![CDATA[about-me]]></category>
            <category><![CDATA[sports]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[basketball-analytics]]></category>
            <category><![CDATA[basketball]]></category>
            <dc:creator><![CDATA[Paul DeVos]]></dc:creator>
            <pubDate>Mon, 15 May 2017 03:29:42 GMT</pubDate>
            <atom:updated>2017-06-26T16:40:47.445Z</atom:updated>
            <content:encoded><![CDATA[<p>One, because I love sports statistics. Ever since I could read or understand numbers and sports I would be on the lookout for any book, journal, or newspaper that might have a boxscore. When it came to sports card where most people were after the “rookie cards” I was after the “last year” cards because it had the fullest data set.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/240/1*mB4GQ5QFSR2VsNAiaUWTgA.jpeg" /></figure><p>When it came to video games — e.g. Tecmo Bowl (Nintendo), R.B.I. Baseball, Double Dribble and Bulls vs Blazers. I would manually enter the stats into a notebook and play full seasons of games to get the stats. It was laborious for sure, but definitely a labor of love. As games evolved, from Tecmo Super Bowl to NBA Live and Ken Griffey Jr Baseball, to keep track of the boxscore stats I would still, at the conclusion of every season or franchise, toss all of these stats into my “databases” [notebooks] as to be able to do further analysis. I did this until I was probably 17 years old and at that time I had hundreds of notebooks.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/550/1*coZXcpeVILNVHb2XyHLU5g.png" /><figcaption><strong><em>NBA Live 1995</em></strong></figcaption></figure><p><em>Sidenote: Do you not find it amazing that pre-teens were doing millions and millions of basketball simulations, aka basketball video games, and could tell NBA GMs, Executives, and Coaches without hesitation that 3-pointers and dunks were the most efficient shots to win games. Amazing. The whole time 10 year old kids knew what took adults 25+ years to figure out.</em></p><p>Two, for a continued learning experience. I love data, as if point number one didn’t say it clear enough, and this is a chance to share that love with others who might enjoy some of the same. It’s also a learning experience. Growing up on a farm in a very rural area I wasn’t able to get firsthand exposure to gets for professional sports teams or even major NCAA teams. I also was not able to get any of those connections to get the opportunities to work in professional sports. So along with those notebooks, <a href="https://evernote.com/"><em>Evernote</em></a> [highly recommended], and a few other places this will serve as a consolidated place for learning more about sports analytics. I will probably focus mostly on basketball as it’s the sport I know most about while also having less known than say baseball does. That bleeds into number three.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/605/1*3GqAGYA74GnMjQxao_vObg.jpeg" /></figure><p>Three, as a place to ‘practice’ data science and collaborate with those who also want to do data science with a focus on basketball. While I would consider myself particularly strong on the “data” side of Data Science I’m still trying to move from ‘able’ practitioner of machine learning to someone who has a deep understanding of the tradeoffs with the various algorithms behind them. I’ve managed to teach myself in a way that I could get employed at Match.com, Google, and IBM Watson as a data engineer — so this hopefully will play a part in the next step in that journey to becoming a better Data Scientist.</p><p>Four, to get better at writing. While many would say I have very good teaching abilities — I taught Astronomy at the college level and AP Calculus at the high school level and won “Most Inspirational Teacher” during my last year of teaching — when it comes to writing to a computer screen that isn’t particular interactive (maybe it’s just my computer) it’s a whole different realm of competency or <em>in</em>-competency in my case. So you’re here and if your ears are itching more than your eyes are bleeding — welcome. I hope it can be a place that’s worthy of your time.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=983c1248b2c6" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Story #1]]></title>
            <link>https://medium.com/@PaulDeVos7/story-1-c95fced7abae?source=rss-608456fc7b77------2</link>
            <guid isPermaLink="false">https://medium.com/p/c95fced7abae</guid>
            <category><![CDATA[test]]></category>
            <dc:creator><![CDATA[Paul DeVos]]></dc:creator>
            <pubDate>Fri, 21 Apr 2017 04:39:08 GMT</pubDate>
            <atom:updated>2017-04-21T04:39:08.222Z</atom:updated>
            <content:encoded><![CDATA[<p>Test. Test. Test.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c95fced7abae" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Testing out publication logo at top]]></title>
            <link>https://medium.com/@PaulDeVos7/testing-out-publication-logo-at-top-4c19b8c9f133?source=rss-608456fc7b77------2</link>
            <guid isPermaLink="false">https://medium.com/p/4c19b8c9f133</guid>
            <dc:creator><![CDATA[Paul DeVos]]></dc:creator>
            <pubDate>Fri, 21 Apr 2017 04:30:59 GMT</pubDate>
            <atom:updated>2017-04-21T04:30:59.217Z</atom:updated>
            <content:encoded><![CDATA[<p>yep yep yep</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=4c19b8c9f133" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Main page Story]]></title>
            <link>https://medium.com/@PaulDeVos7/main-page-story-1da3374f901b?source=rss-608456fc7b77------2</link>
            <guid isPermaLink="false">https://medium.com/p/1da3374f901b</guid>
            <category><![CDATA[data-engineering]]></category>
            <category><![CDATA[python]]></category>
            <dc:creator><![CDATA[Paul DeVos]]></dc:creator>
            <pubDate>Fri, 21 Apr 2017 04:00:12 GMT</pubDate>
            <atom:updated>2017-04-21T04:00:12.399Z</atom:updated>
            <content:encoded><![CDATA[<p>This is the main page story.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1da3374f901b" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[I want this story under Data Science Tab]]></title>
            <link>https://medium.com/@PaulDeVos7/i-want-this-story-under-data-science-tab-a1d351e915c3?source=rss-608456fc7b77------2</link>
            <guid isPermaLink="false">https://medium.com/p/a1d351e915c3</guid>
            <category><![CDATA[data-science]]></category>
            <dc:creator><![CDATA[Paul DeVos]]></dc:creator>
            <pubDate>Fri, 21 Apr 2017 03:57:41 GMT</pubDate>
            <atom:updated>2017-04-21T03:57:41.414Z</atom:updated>
            <content:encoded><![CDATA[<p>etc etc etc</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a1d351e915c3" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Hmmm Try #2]]></title>
            <link>https://medium.com/@PaulDeVos7/hmmm-try-2-5ac2905c2941?source=rss-608456fc7b77------2</link>
            <guid isPermaLink="false">https://medium.com/p/5ac2905c2941</guid>
            <category><![CDATA[data-science]]></category>
            <dc:creator><![CDATA[Paul DeVos]]></dc:creator>
            <pubDate>Fri, 21 Apr 2017 03:18:58 GMT</pubDate>
            <atom:updated>2017-04-21T03:24:08.539Z</atom:updated>
            <content:encoded><![CDATA[<p>This needs to be underneath “Data Science” tab in my “codeblog” publication.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5ac2905c2941" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Data Science Example 1]]></title>
            <link>https://medium.com/@PaulDeVos7/data-science-example-1-9acf93aef7ca?source=rss-608456fc7b77------2</link>
            <guid isPermaLink="false">https://medium.com/p/9acf93aef7ca</guid>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[data-science]]></category>
            <dc:creator><![CDATA[Paul DeVos]]></dc:creator>
            <pubDate>Fri, 21 Apr 2017 03:17:06 GMT</pubDate>
            <atom:updated>2017-04-21T03:34:15.794Z</atom:updated>
            <content:encoded><![CDATA[<p>Python is a great tool for building Machine Learning models.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9acf93aef7ca" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>