2026 Tigers Starting Rotation Projections

The Tigers recently signed Framber Valdez and Justin Verlander, moving their rotation to one of the best projected in baseball. A quick look at their projections from Fangraphs is enough to make any Tiger fan happy.

library(hrbrthemes)
library(Hmisc)
library(janitor)
library(gt)
library(gtExtras)
library(mlbplotR)
library(tidyverse)
library(plotly)

fg_sp_proj <- read_csv('fangraphs-leaderboard-projections.csv') %>%
  clean_names()

fg_sp_proj <- fg_sp_proj %>%
  filter_out(is.na(team))

fg_sp_proj %>%
  filter(team == 'DET') %>%
  select(name, g, ip, k_9, bb_9, hr_9, era, fip, war) %>%
  gt() %>%
  gt_theme_espn() %>%
  fmt_number(columns = 3:9, decimals = 1)

name	g	ip	k_9	bb_9	hr_9	era	fip	war
Tarik Skubal	32	200.0	11.0	1.8	0.8	2.7	2.6	6.3
Framber Valdez	31	195.0	8.3	3.0	0.8	3.4	3.5	3.6
Jack Flaherty	29	160.0	9.9	3.3	1.2	4.0	3.9	2.5
Casey Mize	28	145.0	8.0	2.5	1.2	4.1	4.0	2.1
Justin Verlander	26	141.0	7.4	2.8	1.3	4.3	4.4	1.7

It’s strange seeing Verlander with the worst projection, but 1.7 WAR is pretty good for a fifth starter, especially one of his age*. We could compare to 2011, where Skubal is now Verlander and Verlander is now Doug Fister.

*Since JV is only a year younger than me, we won’t call him old.

But how does this compare to the rest of baseball? First, we’ll visualize the data, then we’ll plot it.

team_plot <- fg_sp_proj %>%
  group_by(team) %>%
  mutate(team_war = sum(war)) %>%
  mutate(team2 = paste(team, round(team_war, 1))) %>%
  ggplot(aes(x = war, y = reorder(team2, team_war),
             color = team,
             fill = team,
             text = paste(
    name, " ", team, "<br>",
    "WAR:", round(war,1), "<br>",
    "IP", ip, "<br>",
    "K/9", round(k_9,1), "<br>",
    "BB/9", round(bb_9,1), "<br>",
    "HR/9", round(hr_9,1), "<br>",
    "ERA", round(era,2), "<br>",
    "FIP", round(fip,2)
  )))+
  geom_point()+
  scale_color_mlb()+
  scale_fill_mlb()+
  theme_ipsum()+
  theme(text=element_text(size = 16,  family="Oswald"),
        panel.grid.minor = element_blank(),
        plot.title.position = "plot",
        legend.position = element_blank())+
  ylab('')+
  xlab('WAR')

ggplotly(team_plot, tooltip = "text") %>%
  layout(title = "Fangraphs Projected Starting Pitcher WAR") %>% 
  layout(showlegend = FALSE)

The Tigers have the second-best projected rotation, led by Tarik Skubal, who is projected to be the best pitcher by WAR this year. (Though the projections seem kinda bearish on Skenes). It’s also worth noting that Verlander’s 1.7 WAR would be highest on several teams, which means the Rockies are in for another rough year.

Now let’s summarize the data. We’ll look at team projected starting pitching WAR, average per pitcher, the standard deviation, and the coefficient of variation (COV), which is simply SD / mean. This helps us get a feel for how consistent a team is relative to it’s mean, with smaller numbers meaning more consistency. A team can be bad and consistent, so we need the average to give us context. We’ll use a weighted average so starters with lower WAR due to 3-4* starts don’t bring the numbers down.

*Or even 6-7.

fg_sp_proj %>%
  mutate(pitcher_war_rank = rank(desc(war)),
         pitcher_era_rank = rank(era),
         pitcher_fip_rank = rank(fip)) %>%
  summarise(pitchers = n(),
            ip = sum(ip),
            gs = sum(gs),
            team_war = sum(war),
            mean_war = weighted.mean(war, g),
            sd_war = sqrt(wtd.var(war, g)),
            .by = team) %>%
  arrange(desc(team_war)) %>%
  mutate(WAR_COV = sd_war / mean_war,
         ip_per_g = ip / gs) %>%
  relocate(WAR_COV, .after = sd_war) %>%
  relocate(ip_per_g, .after = gs) %>%
  gt() %>%
  gt_fmt_mlb_logo(columns = "team") %>%
  gt_theme_espn() %>%
  fmt_number(columns = 3:9, decimals = 1)

team	pitchers	ip	gs	ip_per_g	team_war	mean_war	sd_war	WAR_COV
	8	990.0	155.0	6.4	17.7	2.1	1.7	0.8
	5	841.0	146.0	5.8	16.2	3.4	1.7	0.5
	10	1,005.0	162.0	6.2	16.1	2.1	1.7	0.8
	9	927.0	156.0	5.9	16.1	2.0	1.1	0.6
	7	878.0	155.0	5.7	14.5	2.3	1.0	0.4
	9	995.0	158.0	6.3	14.5	1.7	1.1	0.7
	7	956.0	156.0	6.1	14.2	2.1	1.0	0.5
	8	948.0	158.0	6.0	14.1	1.9	1.5	0.8
	9	946.0	158.0	6.0	13.7	1.8	1.1	0.6
	6	860.0	151.0	5.7	13.7	2.7	0.8	0.3
	7	945.0	153.0	6.2	13.5	1.8	0.9	0.5
	8	988.0	159.0	6.2	13.4	1.6	1.0	0.6
	8	927.0	154.0	6.0	13.2	1.8	1.2	0.7
	9	940.0	157.0	6.0	12.9	1.7	0.8	0.5
	7	904.0	155.0	5.8	12.0	1.8	0.6	0.3
	8	948.0	159.0	6.0	11.8	1.5	0.7	0.5
	8	927.0	158.0	5.9	11.4	1.5	0.7	0.5
	7	903.0	156.0	5.8	11.4	1.8	0.5	0.3
	10	900.0	160.0	5.6	11.0	1.7	1.0	0.6
	9	967.0	158.0	6.1	10.8	1.3	1.0	0.8
	7	894.0	156.0	5.7	10.8	1.5	0.6	0.4
	9	945.0	160.0	5.9	10.7	1.5	0.7	0.5
	8	937.0	157.0	6.0	10.7	1.5	1.4	0.9
	9	939.0	154.0	6.1	9.9	1.2	0.7	0.6
	8	930.0	160.0	5.8	9.8	1.5	0.6	0.4
	10	937.0	163.0	5.7	9.7	1.4	1.2	0.8
	10	982.0	163.0	6.0	8.8	1.1	0.5	0.4
	9	902.0	155.0	5.8	7.9	1.0	0.6	0.6
	8	886.0	148.0	6.0	7.4	1.1	0.5	0.5
	10	938.0	160.0	5.9	6.7	0.7	0.5	0.7

What’s interesting here is that, despite having the second-highest projected WAR in MLB, the Tigers are only projected to throw 841 innings, the lowest in the game. The depth charts only have 146 games slotted, so they’re not fully taking depth into account. The ~6 innings per game average is concerning, though, since their bullpen projects in the bottom half. Bullpens are hard to project, so we can hope that they’re better here.

*Yes, IP per game started looks funky. I didn’t convert to outs, so it’s a bit off.

But how much is having only the top 5 in here impacting the IP variable? We know that we’ll need our depth at some point, either in a double header or to sub while someone’s on the IL. It’s strange not seeing Montero, Melton, or Anderson here (and Jobe may come back late season), so I suspect there’s something amiss with how FG compiled this data. In fact, if we compare IP per starter to the rest of the league, we see a clear outlier:

fg_sp_proj %>%
  summarise(war = sum(war),
            ip = sum(ip),
            n = n(),
            .by = team) %>%
  mutate(ip_per = ip / n) %>%
  ggplot(aes(x = war, y = ip_per))+
  mlbplotR::geom_mlb_logos(aes(team_abbr = team), width = 0.055)+
  theme_ipsum()+
  theme(text=element_text(size = 16,  family="Oswald"),
        panel.grid.minor = element_blank(),
        plot.title.position = "plot",
        legend.position = element_blank())+
  ylab('Projected Mean Innings Pitched Per Started')

So, I’d take this to mean that the Tigers should have one of the best rotations in the game this year, even assuming we’ll need our depth guys at some point. The bullpen is still a concern, but as a lifelong Tiger fan that’s just an annual thing. The bats are a bit of a question yet - I think they’ll be good, but not great - so the pitching doing its thing is going to be a big component of their success this year *.

*It would’ve saved time to just say this and call it good, but alas.