Close Menu
    Daten

    [OC] Ich habe alle 68 March Madness-Teams danach eingestuft, wie viel sie tatsächlich die Löhne erhöhen – bereinigt um Lebenshaltungskosten, Abbrecherquoten und Schulzugänglichkeit. 20 Teams haben einen NEGATIVEN ROI.


    [OC] Ich habe alle 68 March Madness-Teams danach eingestuft, wie viel sie tatsächlich die Löhne erhöhen – bereinigt um Lebenshaltungskosten, Abbrecherquoten und Schulzugänglichkeit. 20 Teams haben einen NEGATIVEN ROI.

    Von adilkhan1214

    View 7 Comments

    7 Kommentare

    1. adilkhan1214 on

      **Interactive version:** [https://adil1248.github.io/march-madness-roi-2026/](https://adil1248.github.io/march-madness-roi-2026/)

      **What this shows:**

      I took all 68 teams in the 2026 NCAA tournament and asked: does attending this school actually make you richer than just entering the workforce with a high school diploma?

      The twist: instead of using raw salary numbers, I adjusted for three things most rankings ignore:

      1. **Regional cost of living:** A $55K salary in Iowa has more purchasing power than $65K in New York. I used BEA Regional Price Parities to convert everything to real dollars, and compared regional schools against their local HS baseline, not the national one.
      2. **Graduation rates:** If a school has a 35% grad rate, 65% of enrollees are paying tuition for a credential they never receive. Those dropouts get zero earnings boost but still lost money. This can make the adjusted ROI deeply negative.
      3. **Accessibility & scale:** A school with 80% acceptance and 50K students delivering solid ROI is arguably more impactful than a 5% acceptance school with amazing ROI for its 3,000 students. The „Impact Score“ weights for this.

      **Key findings:**

      * 20 of 68 teams have **negative adjusted ROI** graduates literally earn less than local HS-only workers
      * Large accessible state schools like **Purdue** dominate when you factor in scale
      * Several elite schools (Penn, Duke, Vanderbilt) have incredible per-student ROI but low Impact scores because they’re tiny and ultra-selective
      * Low-COL states like Iowa, Kansas, and Arkansas get a big boost: modest salaries go much further there
      * The biggest value destroyers are schools with <40% grad rates in states where the HS baseline is already decent

      The idea, methodology, design, layout, and initial hypothesis investigation was all done without Ai, the visualization/html/UX was created with Claude Code. This allows me to quickly update/iterate if there is any challenges to the methodology or a missing insight within minutes. I believe this to still be an original presentation of data that nobody else has done.

      **Source & Tools:**

      |Data|Source|
      |:-|:-|
      |Earnings|U.S. Dept. of Education College Scorecard (median earnings 10 years after enrollment)|
      |Regional cost of living|Bureau of Economic Analysis Regional Price Parities, 2024|
      |HS baseline|Bureau of Labor Statistics CPS Q1 2025 ($50K/yr national median for HS-diploma-only workers)|
      |Graduation rates|IPEDS 6-year rates|
      |ROI framework|Georgetown CEW „Ranking 4,600 Colleges by ROI“ (2025)|
      |Acceptance rates & enrollment|Common Data Sets 2024-25|
      |Built with|HTML/JS, hand-compiled data|

      **GitHub repo with full methodology:** [https://github.com/adil1248/march-madness-roi-2026](https://github.com/adil1248/march-madness-roi-2026)

    2. Suspicious_Funny4978 on

      The methodology here is solid, but the negative ROI numbers probably hide a harder truth: prestige schools stay prestigious even when they should have crater in college selection rankings. Parents and students follow brand signals that are decades out of date.

      What jumps out is that some of the lowest ROI schools are probably the ones that most aggressively market themselves as life-changing. The athletic schools especially—you get the sunk cost of four years, plus the student-debt model, plus a degree that doesn’t produce the promised earnings.

      The interactive version would be way more useful if it surfaced which schools have the highest gap between brand perception and actual outcomes. That’s where the real arbitrage is.

    4. SomethingMoreToSay on

      This is interesting, but I have to question one aspect of your methodology.

      You say that

      >20 of 68 teams have **negative adjusted ROI** graduates literally earn less than local HS-only workers

      but that’s not what you’re calculating in your „adjusted ROI“. You’re factoring in the drop-out rate, so what you mean is that the average *student who goes to that university* earns less than high school graduates. But the average *graduate of that university* could well earn more.

    5. nian2326076 on

      For interview prep, especially if you want to stand out, focus on showing how your experiences match what the employer values. Tailor your answers to highlight skills and achievements that fit the job description. Practice common interview questions, but don’t memorize them. Instead, have key points you want to cover. Be prepared to talk about how you handle challenges, as that often comes up. If you need a resource for mock interviews or feedback, I’ve found PracHub helpful. They have some good tools for practicing with others and getting constructive criticism. Good luck!

    6. ArkGuardian on

      This methodology is fine but there might be some hidden variables here.

      Especially since Howard is listed as very negative ROI and has markedly different demographics then some of the key industries near it

    7. T0WER89 on

      Basically all the top “impact” schools are just very large universities. I agree that Purdue helps more people than Duke and that should be accounted for but the top of this data set are the largest universities in the tournament.

    Leave A Reply