| Course | MT241 Sport Analytics |
|---|---|
| Unit | Unit 6 |
| Paper type | Seminar reflection |
| Length | About 820 words, 5 pages double-spaced |
| Format | APA 7 student paper |
| School | Purdue University Global |
| Program | BS in Business Administration |
| Updated | October 2026 |
Free sample paper for MT241 Unit 6
Thirteen Points of Save Percentage That Shrank to One Goal: A Unit 6 Seminar Reflection on Two Goaltenders
[Student Name]
Purdue University Global
MT241: Sport Analytics
Unit 6 Seminar Reflection
[Instructor Name]
[Date]
The club, both goaltenders, the file and the classmates are composite and illustrative; the sources cited are real.
Where I Started
My plan going in was to argue that our composite club should start its backup. The numbers in the file looked plain. Across all situations, the starter had stopped 90.8 percent of the 1,655 shots he faced, and the backup had stopped 92.1 percent of 494. Thirteen points of save percentage sounded large to me. Over a thousand shots it is thirteen goals, and thirteen goals is roughly two wins in the standings. I wrote my discussion notes around that difference and expected the seminar to be about whether the coach was being stubborn.
The First Trap: How Much 494 Shots Can Say
The first question came from our instructor, who asked how sure we were that the backup's number would hold up over another 494 shots. One classmate, Priya, worked out a rough 95 percent interval for each figure during the session. For the starter, with 1,655 shots, the interval ran from about .894 to .922. For the backup, with only 494, it ran from about .897 to .945. The two ranges overlap across most of their width. In plain terms, the data are consistent with the backup being better, about the same, or slightly worse. The smaller sample makes his range much wider, and nearly all of the starter's range sits inside it.
I had known in a general way that small samples are noisy, but I had not thought of save percentage as a rate built on a fairly small number of events. Agresti and Coull (1998) show that simple intervals for proportions like this one can be too narrow when samples are small, which would make the overlap even larger than Priya's estimate.
The Second Trap: Not All Shots Are Equal
The second trap was quieter, and I would not have found it on my own. Another classmate, Jordan, pointed out that the backup's starts were not a random sample of games. Most of them fell on the second night of back-to-back games, when coaches rest their starters, and the schedule had given him mostly weaker opponents on those nights. The file includes expected goals for each shot, and when we added them up, the backup had faced 0.071 expected goals per shot and the starter 0.083. The starter had been facing harder shots.
The way to compare them fairly is to measure goals saved above expected: how many goals each goaltender allowed compared with how many an average goaltender would have allowed on the same shots. The public explanation on MoneyPuck (n.d.) sets out this method. When we did that, both goaltenders came out slightly below average, and the difference between them was about one goal per thousand shots. The thirteen-point gap had almost disappeared.
The backup was not stopping more pucks; he was facing easier ones.
What Changed in My Thinking
What changed was not my view of the backup as a goaltender. He may be good. What changed was my sense of what the file could tell us. I had read save percentage as a measure of skill, when it is also a measure of the shots a goaltender happens to face and of how many of them there were. Alamar (2013) describes the analyst's job as separating what a number captures from what we hope it captures, and I had skipped that step because the number agreed with what I already thought.
I also noticed that both traps pointed the same way. A small sample and an easier schedule each made the backup look better than the evidence supports. When two problems push in the same direction, the conclusion they support deserves extra doubt. Had the backup faced harder shots, the small sample would only have made his number uncertain; here it also made it flattering. I want to remember that pattern for the player comparisons in later units, where usage and sample size can stack up in the same way.
Where I Ended Up
Benching the starter, I now think, is not something this file can support. I also do not think it shows that the starter is better. The honest conclusion is that this season's numbers cannot separate the two goaltenders, and the decision should rest on other evidence: the coaches' judgment of their play, their health and their workload. If the club wants to keep using data, the useful step is to give the backup more starts against ordinary opponents and review the comparison after another few hundred shots. That is less exciting than the argument I came in with, but it is the one I can defend.
References
Agresti, A., & Coull, B. A. (1998). Approximate is better than "exact" for interval estimation of binomial proportions. The American Statistician, 52(2), 119-126. https://doi.org/10.1080/00031305.1998.10480550
Alamar, B. C. (2013). Sports analytics: A guide for coaches, managers, and other decision makers. Columbia University Press.
MoneyPuck. (n.d.). About and how it works. https://moneypuck.com/about.htm
What the MT241 Unit 6 assignment asks for
The sixth unit's seminar works through a data set as a group, and the reflection asks what you thought before, what the session showed and where you ended up. Name the specific evidence that challenged your view and who raised it, using first names or roles only. The prompt expects course concepts such as sample size, intervals or adjusting for difficulty to be applied to the case, with at least two sources in APA 7. If you missed the live seminar, the alternative assignment may require answers to the seminar questions as well, so check your section's instructions. Two pages of first-person writing is typical, and a small table of the figures discussed often helps a reader follow the change of view.
How this MT241 Unit 6 example is built
The reflection starts with the writer's plan and the arithmetic that made thirteen points of save percentage sound like two wins. The first trap, sample size, records Priya's intervals and why they overlap, with Agresti and Coull on small samples and a margin note on the evidence that changed the view. The second trap records Jordan's point about back-to-back starts and the per-shot expected goals, then applies goals saved above expected, and the highlighted line says the backup faced easier shots. A section on what changed applies Alamar's distinction between what a number captures and what we hope it captures, and adds a note on two errors pointing the same way. The last section states the revised position and a plan for more starts.
Where the points sit in the MT241 Unit 6 rubric
Reflections in MT 241 seminars are marked chiefly on honest reasoning and correct use of analytics concepts. Credit goes to stating a clear starting position, naming the evidence that tested it and explaining the change, or the reasons for keeping the view, in your own words. Concepts earn marks when they are applied to the figures discussed rather than defined. Graders reward reflections that draw a general lesson beyond the single case. First person is expected, but figures and methods still need sources. A short table of the figures discussed, accurate APA references and a closing position the writer can defend complete the rubric, and a reflection that keeps to the two-page limit is easier to mark than one that runs long.
MT241 Unit 6 help from the desk
A common weakness is retelling the seminar in order without showing what the writer thought at each point. Another is saying the seminar changed your mind while leaving out which figure or comment did it, which makes the change impossible to judge. Some students write intervals or expected-goals figures that do not match the data discussed. Others end with a strong new recommendation when the honest conclusion is that the data cannot decide. Keep classmates' names to first names. If your seminar used a different data set, our writers can help you shape a reflection around the figures your group actually examined.
Get an MT241 Unit 6 example written to your instructions
Goaltending numbers, player ratings, a team's shooting file: whatever the Unit 6 session put in front of the group, outline it along with the view you held beforehand. Say whether the reflection substitutes for live attendance, and attach the rubric. The first custom reflection is free, returned in 24-48h, with the dataset on the page and each trap named. The paper above is an original model document written by our desk, not a submitted student paper and not an official Purdue University Global document.
More MT241 papers and BS in Business Administration samples
- MT241 Unit 1: Discussion Board Post
- MT241 Unit 2: Metric Definition Exercise
- MT241 Unit 3: Descriptive Analysis
- MT241 Unit 4: Player Comparison Report
- MT241 Unit 5: Data Quality Review
- MT241 Unit 7: Visualization Exercise
- MT241 Unit 8: Tactical Analysis
- MT241 Unit 9: Recruitment Recommendation
- MT241 Unit 10: Analytics Report
- MT240 Unit 6: Seminar Reflection
- MT237 Unit 6: Seminar Reflection
- MT236 Unit 6: Transportation Comparison
- MT221 Unit 6: Service Process Map
MT241 Unit 6 questions, answered
How big a sample does a goaltender's save percentage need?
Larger than one season usually provides. Save percentages cluster tightly across the league, so the differences that matter are small, and random variation over a few hundred shots exceeds the gaps between most goaltenders. An interval, even a rough binomial one, shows how much of a difference could be noise. The example's intervals overlap almost completely.
What is goals saved above expected?
The difference between the goals a goaltender would be expected to allow, given the quality of the shots faced as scored by an expected-goals model, and the goals actually allowed. A positive figure means he stopped more than the shots predicted. It adjusts for shot quality, which raw save percentage ignores, though it inherits the model's blind spots, such as screens and pre-shot passes.
Can a reflection end without a firm recommendation?
It can end with a firm position about the evidence instead of a verdict on the players. The example keeps the starter because the data does not justify a change, then sets a date and a threshold for revisiting. That is a decision with its conditions stated, and it usually earns more than a confident choice the numbers cannot support.
Where can I find a free MT241 Unit 6 sample paper?
This page carries the full MT241 Unit 6 seminar reflection. It follows a student who favored a backup goaltender on save percentage until intervals and shot difficulty, raised in seminar, closed most of the gap.
What two traps does the MT241 Unit 6 sample describe?
A small sample, 494 shots, which makes the backup's save percentage uncertain, and easier shots, since his starts fell mostly on back-to-back nights against weaker teams. Both made him look better than the evidence supports.