 Hi, I'm Geoff Watts and welcome to episode 2 of Scrum Mastery Challenge, the Ultimate Agile Coach Game Show. In our first episode, which was an artistic challenge, Christina was judged to have produced a picture that most represented an aspect of agile delivery. Now that challenge tested their abilities to interpret, empathize and get their message across to others. Today's challenge will test different abilities, but abilities that are arguably equally important in an agile context. Let's see what this episode has in store for our competitors. Estimate the number of pieces of pasta in the jar. Close this estimate win. Estimation is a tough challenge and all agile teams will come up against it at some point. The good news is that here we've got some experts. Some agile coaches with years of experience of estimation and a lot of the theory behind it. An estimating pasta isn't a win near as complicated as estimating the difficulty of real multi-faceted work. So this challenge should be easy, right? And I'm assuming I'm not allowed to open it. Well, what does the task say? Okay, estimate. Yeah. Bad at numbers. 150. 150. It's kind of tough. It's sort of like thin pasta. And so they, and it's got like gaps so they can sort of like mesh and then it saves space. But then they're quite long. Yeah, I think about 200. 4, 5, 6, 7, 8, 11. Actually, no, I'm not because that's a very precise number. I'm going to go 320. That is my answer. Go on with 353. What's the thought process then? I could see about space for about 20 at the bottom. And then it seemed like there was about sort of 20 layers worth of pasta. So I've sort of assumed that there's some air in there. Um, gone 350. 20 times 20 is 400. But we're even terrible at estimating large numbers. See, there's a theory, isn't there? 123. I'm going to say, Jeff, 148. 150 initially. But then I thought, I counted a couple of those. I thought, well, no, probably weren't how 150. So I went to around 90. Comparing the two jars, maybe just under double. So I put a number of 160. All right, 450. They're the same as those, but they're some that are a bit different and they're thinner. But then the jars are a lot bigger. So I thought, obviously more are going to fit in. Um, yeah, I reckon it's just under double the size. So hundreds. I mean, the fact that you've put different types of pasta in here is just, that's completely all better off. I've just brought your pen, Jeff. Never mind. 718. Yeah, 7, sorry, 780. Yeah. Can I look at both of you? Right, does it give me a clue numbers of how big the jars are? It doesn't, does it? 20% larger. So I'll just do a 50%. 300. And yeah, so keep it, keep it simple. 300 and let's say 300. Change again. So why? Well, I counted how many bits of pasta there were in this section, and I was kind of just randomly guesstimated and sort of said, right, there's about 10 here, 10, 20, 30, whatever. So I figured that there's actually more like 80 in here. What's really funny was I put 150 on this originally, and then I changed it because I started doubting myself From 80, this looks like it's roughly double. That went back to 150. And this one, it doesn't look like it's double. I think a bit less, so about 200. But 70. Okay. Because they're all the thin ones, I can put a few in there. So this one is much smaller. I'm going to go 100. Okay, bring it on. Jar three. A bit less than that, look quite well. Probably about, looks like about a third of that job. I mean, I'm pretty much, my strategy now is based on the fact that this is close to correct. Different pasta, Jeff. What was it? No, no, no, no. There's different pasta in that one now. I'm making an assumption, you see. I say it's less, it's about half, but just more than half. So I'll say, I'm going to say 74. You estimated 50. Happy? Found a penny today, if luck might be on my side. Yes, I'm happy. Okay. We're going for 49. 49. So there we have our estimates. Some very big differences, but much more variation on the bigger jars than the smaller ones. When given the chance to change their minds one final time, Christina and Freya both changed their estimates, but Paul, Sam and Helen stuck with their originals. Let's see how they did. 298. Oh, blimey, really? Yeah. Oh, really? Are you sure? Does it not look right that many? That's not bad. All right, I'm actually quite comfortable with that. It doesn't. Oh, I'm even close. Okay, so I was more out of that. Oh, yes, 718. Really? Oh, blimey. Okay. Well, it wasn't too bad. I mean, this was really far off, but this one was too bad. Okay. Whoa. Do you know what? That is so annoying. I was at 120, wasn't I? Three. All right, I'm good. I don't even want to know this one now. Let me tell you there are 53. Oh, that's not bad. Okay. I'm quite happy with that. Happy with that. Yeah. 53. Okay. Oh, it's tiny, so tiny. I can probably count these. Can I open it? Can I open it? What does the task say? Yes, I can open it. There's nothing stopping me opening it. 53. Don't break the rules. Did you know? I didn't know, did I? It doesn't say I can't open the jars. What does it ask you to do? Estimate the number of pieces that pass through the jars. Did you estimate the number of pieces that pass through that jar? No, I didn't. As you can see, Sam stormed this challenge, and is probably now looking out for a new career as a quantity surveyor. Helen was also rightly pleased with her performance, while Freya and Christina were very surprised how far out their estimates were. And Paul, well, the less said about him, the better, I think. But was there anything informative about what happened here? The first interesting observation for me here was around what we call orders of magnitude. The smaller the jars are past there, the smaller the range of estimates. And the more confident our contestants were in their estimates. This is because the numbers involved are easier to envisage and deal with. And as you can see on this table, there's a 55% range on the biggest jar compared to just a 29% range on the smallest jar. For me, the obvious takeaway here is that if you want better estimates, then give people smaller pieces of work. The second observation for me was that our competitors all gravitated towards relative estimation. Now, it's probably not a surprise given that our contestants are agile coaches, but even Freya, who isn't, found comparing much easier than an actual estimation. You'll have noticed our contestants picking the jars up and looking to weigh them up against each other. The beauty with relative estimation is that if we're relatively right, then we can quickly take advantage of any actual data that we can get to make all of our estimates more accurate. And if we're relatively wrong, then it's much quicker to re-evaluate our estimates. Freya was relatively right in her comparisons, even though her actual numbers were wrong. And once she has some actual data about her imprecision, the correct ratios that she had would be very useful for her. There was a difference between the amount of time and effort that some of our contestants put into estimating compared to others. And it seemed to me that the people who put the most amount of effort into trying to be precise ended up with the most furthest-out estimates. Perhaps we can overthink things and we can actually work ourselves away from the right answer. Interestingly, people found it difficult to change their estimates. Sometimes, even when they received data that suggested their initial estimates were wrong. By giving contestants smaller jarred later, they were able to come up with estimates that they were more confident about. Then logically, they should be able to use that data to adjust their bigger estimates from earlier. But not all of them did that, because they were anchored to their previous estimates. Even when Paul actually counted the pasta, naughty Paul, and he had real data, he was still overly attached to his original estimates. For me, this strongly suggests we should start our estimation process with the smaller items, because these are less likely to be wrong. The final interesting aspect for me here was nothing to do with the actual estimation process, but actually related to what we call confirmation bias. Both Christina and Paul wanted to count the pieces of pasta in the smallest jar. I told them both to reread the task, to see if that was allowed, and they both saw different things, even though the wording was exactly the same. As Annay Ninh once said, we don't see the world as it is, we see it as we are, and we tend to look for what we expect, and what we want to see, rather than what's actually there, and we'll filter out what we don't want to see or don't expect to see. In general, I'm in favour of challenging rules and assumptions to be creative, but as Paul learned the hard way, you've got to do that carefully. Estimation is a difficult thing for people to do. Generally speaking, human beings are not very good at estimating, and when we're doing it on our own, we've only got our opinion to base things on. Luckily, in an agile team, we can take advantage of lots of different perspectives, people with different experiences, people who are optimistic, pessimistic, they can see different angles to things, and weigh those up almost in a wisdom of crowds kind of approach, come up with something more useful and rounded as a team. So let's see what that does to the leaderboard, now that we're two challenges in. Well, Paul's misreading of the challenge leaves him last, while Christina maintains her overall lead, albeit only just, with a bunching up in the middle, as Sam, Helen and Freya are all on six points. This means it's very close as we move forward with the competition. Well, that's the end of our time box for episode two of Scrum Mastery Challenge, and we're almost done. I hope you enjoyed the episode and found it useful. Please make sure you're subscribed to the YouTube channel, so you'll be the first to know, and you don't get any annoying spoilers from your colleagues at work. Leave us some comments. What would you have estimated? Do you have any good ways of estimating in your teams? Until next time, good luck with your own Scrum Mastery challenges, and to get us to Dun Dun, here are the credits.